Mar 9, 2021
- 3:10 - The humble Hugh Millen
- 12:40 - Hugh's data origin story heavily involves football and his data knowledge parlayed into big bucks
- 28:30 - The impact of data in sports and the parallels of Sports and Business analytics
- 57:20 - A discussion about the value of football in society yields some deep and philosophical tangents
Rob Collie (00:00:00):
Hello, friends. I'll be honest, what sorts of stereotypes come to
mind when I say the words, professional athlete. We'll up the ante
and say, what about professional NFL football player? You probably
don't think data nerd when you hear those phrases do you? Yet 10
years ago, when I was in Seattle on business visiting Microsoft, I
went on this cloak and dagger side mission one night, drove out
into the foothills of the Cascade Mountains. It was a very small
town, with one stoplight, I parked my car in what was essentially
like a logging bar, like a bar where vloggers would hang out and
drink at night and that's when I met this week's guest, Hugh
Millen. He walked into this joint with a gigantic laptop under his
arm, and sat down with his back against the wall, so no one could
see what was on his screen.
Rob Collie (00:00:54):
Then, he showed me what was the most amazing spreadsheet I have
ever seen. I won't give away what was in that spreadsheet, but I
will tell you, this was the first and last time that I've ever seen
the arc tan, arctangent function used in the wild. He was the real
deal. He was legitimately an NFL star in his day and he loves data
and that's why we've been friends for more than a decade now.
Whether you're into sports or not, whether you're into football or
not, listening to the way his mind works and the conversation that
we had about that domain, I think is still incredibly relevant to
the things that we do in the business space, and so many times
during this conversation, those parallels just kind of jumped off
the page at me, I hope you get as much out of it as we did. Tom
really enjoyed himself on this one in particular. So let's get into
it.
Announcer (00:01:48):
Ladies and gentlemen, may I have your attention please.
Announcer (00:01:52):
This is the Raw Data by P3 podcast with your host, Rob Collie and
your co host, Thomas LaRock. Find out what the experts at P3 can do
for your business. Go to powerpivotpro.com. Raw Data by P3 is data
with the human element.
Rob Collie (00:02:10):
Welcome to the show. Hugh Millen, how are you today?
Hugh Millen (00:02:14):
I'm doing real well guys, how are you?
Rob Collie (00:02:17):
Fantastic. This has been sort of a dream of ours for a while to get
you on this show and it's an honor to have you here. Seriously, I'm
really, really pleased.
Hugh Millen (00:02:24):
Well, thank you. I'm flattered. You're obviously uneducated about
my career because you wouldn't make those statements if you do,
what a hack I have been through the bulk of my career but I'll take
a nice compliment.
Rob Collie (00:02:38):
Yeah, I hear you, so not many people in the world can say that they
have competed in anything at the absolute highest levels that the
planet has to offer and you have. I love the humility about your
career, but one of my favorite sayings, I don't know if you know
this, I have repeated a sentence that you said to me years ago. I
have repeated it a million times since then to other people.
Hugh Millen (00:03:05):
I didn't know that.
Rob Collie (00:03:06):
I don't know if it's a regular line for you or if you just ... I
suspect it's one of your go to lines, but you said to me one time
like, "Hey, you know, I might not have the best NFL career or
whatever," but then you turn, you looked at me and said, "But you
got to beat out an All American for your chance to suck."
Hugh Millen (00:03:23):
Yeah, right. Yeah, you got to beat out an All American in a first
round draft pick to get the chance to suck and I've done all
three.
Thomas LaRock (00:03:32):
That's awesome.
Hugh Millen (00:03:33):
Yeah, that's true.
Rob Collie (00:03:34):
That's an important ... the reason why I repeat that line to people
is because like, I do, I run into people all the time, including
myself who are ... People who are executing at whatever it is they
do at a relatively high level or sometimes an extremely high level,
and you're still going to hit adversity. It's not going to be
peaches and cream. So, a lot of people, especially like in our line
of work suffer with imposter syndrome. When they hit that
adversity, even though that they're actually doing very well,
right, they hit something and they hit a failure, they take it very
personally and it's like invalidating of their whole narrative,
their whole life story. You shouldn't be that way. That's when I
tell them the story. It's like, "Look, essentially they have, in
our world, they've beaten out the All American or the first round
draft pick or whatever and now they've had their chance to suck,"
right? Okay, fine, but what do we do now, right?
Hugh Millen (00:04:28):
Sure. Right.
Rob Collie (00:04:28):
You got to have perspective and that balance perspective about your
background is something that I have immense respect for.
Hugh Millen (00:04:36):
Well, thank you.
Rob Collie (00:04:37):
Most people, they tend to err on one side or the other. Most people
will either say, I'm the greatest and I got screwed over or
something, right? They'll do everything that they can to
rationalize their great self narrative or they will positively
trash themselves. It's hard to be in the middle. It's hard to have
that perspective. It's hard to balance the two and so, this is one
of those little nuggets of wisdom that I like to think that I go
around harvesting and I've used that line so many times.
Hugh Millen (00:05:08):
Good. Good. Thank you.
Rob Collie (00:05:10):
I'm enriched.
Hugh Millen (00:05:10):
Yeah. Well, it's a little bit of wisdom. I think, I ... early on I
had grown up in Seattle, the University of Washington was my
favorite football team. They were the local team and they had
played in Rose Bowls prior to me playing for them and in my junior
year, we started off, we were eight and O. We beat some good teams.
Michigan, for example, in the big house. If you're a college
football fan, you kind of know that. They were ranked number two in
the country. Anyways, we're ranked number one and then about a
month later, I had a really bad first half, weather was a part of
it but I just totally stunk and I got benched at halftime and I was
getting booed in Husky Stadium, and it was so painful that driving
home from the stadium, I pulled off into a Safeway parking lot, a
grocery store parking lot and both my arms were folded over on my
stomach, and I just was kind of like rocking back and forth.
Hugh Millen (00:06:01):
The pain was in my gut and at that time, one of the inspirational
quotes for me was the Teddy Roosevelt, the Man in the Arena.
Rob Collie (00:06:10):
Yeah.
Hugh Millen (00:06:11):
I think it would apply to people in tech world and in all works.
It's worth a look, if you can Google it. I don't have the script in
front of me, I'd butcher it, so I don't even want to try but the
essence of it was, there's few people, who are aspiring to do
really great things but there are many people who want to be
critics and it's really easy to be a critic or unfortunately, the
Bible tells us about the seven deadly sins, pride and envy,
particularly envy, I think most people, all of us were afflicted by
that a little bit. So a lot of times when you're trying to write
this killer code and do something, in whatever profession that
you're doing, if you're trying to be a high achiever and really
soar, and fly with the Eagles, there's somebody who sees you on the
ladder above them and that makes them uneasy, and they want to grab
your ankles and pull you down to their miserable level. I wish I
didn't have to learn that lesson when I was 20. I don't know how it
shaped me.
Rob Collie (00:07:13):
It's a hard thing at that young age in particular.
Hugh Millen (00:07:16):
Yeah.
Rob Collie (00:07:16):
Like you've had the experience of 80,000 people cheering for you,
but you've also had the experience of 80,000 people booing you.
Hugh Millen (00:07:23):
Yeah, yeah.
Rob Collie (00:07:24):
That is intense.
Hugh Millen (00:07:26):
Yeah, it's tough. You got to be mentally strong and I think just
through the adversity, I think for me, I walked on, I didn't get a
scholarship until I earned it later, I had to go to a junior
college, so it was a circuitous route, even to be a college
football player, let alone to the NFL and I tell my kids now who
are aspiring quarterbacks, I just said, "You know, looking back on
it, I never got discouraged." Even through the elements that
discouraged, I might say, "Okay, you tell me ..." and I'm not
saying straight to the face but the evidence is that maybe I suck
now but that's okay. I'm not going to suck tomorrow,
metaphorically.
Thomas LaRock (00:08:01):
Yeah.
Hugh Millen (00:08:01):
I had this vision of, "Okay in 12 months or 18 months or whatever,
I'm going to be a different player then and maybe you're right
now." I had a coach in my high school, I was all league, led the
league and everything, first team all league, kind of honorable
mention in all state. So I was a decent high school player, far
from any John Elway or aggressively recruited guy, but there was a
school, Eastern Washington University, which is Big Sky division
two. Not the Washington Huskies or Washington State Cougars that
you may know who play in the Pac-12. I'm talking about Eastern
Washington. The coach came out, spent the entire day at my high
school watching the tape. When I met him after the afternoon, he'd
been in there for six hours, he goes well ... he seemed to take
glee, looking me right in the eye and say, "You're not the caliber
of player we're looking for at Eastern Washington."
Hugh Millen (00:08:49):
He seemed to really like relish saying that to a 17 year old and
just kind of twisting the knife. I remember just being ... it stung
but I just like, "Okay, maybe you're right now, but you won't be."
Anyway, I think we all kind of go through some challenges like
that, if we're trying to ... I don't want to say cheap greatness
but if you're trying to do something that's competitive and a real
challenge, and really be rare in the world, I think that you're
going to encounter a lot of those type of setbacks.
Rob Collie (00:09:20):
The Roosevelt thing, it starts off with, "It's not the critic who
counts." It's not often that you get Hugh Millen and Brene Brown
bingo, in the same conversation. I mean, that's a big theme of her
work and it's something that I'm a big believer in, our whole
family is a big believer in, like you're going to have naysayers,
like our company's business model for instance, and I won't belabor
the point. I've said it so many times on this show, but we're doing
things right now that a number of people told us, respective people
told us was impossible and we had to ignore that, to go do it but
the criticism still weighed. I personally still carried those
criticisms like I really ... it hurt in a way that people didn't
believe, we've proven them wrong now.
Hugh Millen (00:10:10):
Sure. Yeah.
Rob Collie (00:10:10):
Okay, I'm not going to look back on this, with a different
perspective than I had at the time, but you and I, we didn't cross
paths because of the NFL. I didn't even make my flag football team
in intramural college. I actually retired from intramural flag
football in college, after a pick six.
Hugh Millen (00:10:31):
You threw the pick six?
Rob Collie (00:10:32):
I caught it.
Hugh Millen (00:10:34):
You went out on glory. Okay.
Rob Collie (00:10:35):
I caught it and I ran out and I just basically like kept running.
It's like, leaving on a high note.
Hugh Millen (00:10:41):
You are Bo Jackson and the kingdom, right?
Rob Collie (00:10:43):
That's it, right? Up the tunnel and that was it. My teammates were,
"You're just going to score pick six and retire?" I'm like, yeah.
That's exactly what I'm going to do. Why would I spoil that?
Hugh Millen (00:10:54):
That's it. Well, your blessings were in other areas.
Rob Collie (00:10:56):
So we've known each other for about 10 years.
Hugh Millen (00:10:58):
Yeah.
Rob Collie (00:10:59):
Another thing that I've always enjoyed about knowing you is that we
can take like a year off or even more sometimes without ever
talking to one another, and then when we do start talking again,
it's like we resume mid sentence.
Hugh Millen (00:11:10):
Yeah, good point. Yeah, and that's what happens with friends,
right, is if you can do that, no, yeah, we made that connection,
talking about data and you have a unique affinity to data and
presentation of data and then also football, and you come out
professionally from the data world and you have this interest in
football. I come at it professionally from football who had an
interest in data. So we kind of somehow met in the middle.
Rob Collie (00:11:38):
So you know that saying that all rappers want to be athletes and
all athletes want to be rappers?
Hugh Millen (00:11:44):
Yes.
Rob Collie (00:11:44):
I want you to be the beginning of a new trend, where all nerds want
to be athletes and all athletes want to be nerds. Can we make that
a thing?
Hugh Millen (00:11:53):
I'm a self professed nerd and geek. I call myself that all the time
on the sports radio that I do. I don't find it disparaging at all.
I'm old enough, I can remember a world without Bill Gates. So nerds
were thought to be just kind of interested in things that will
never ... help the world or themselves and then all of a sudden,
all the tech billionaires came around and nerds and geeks, they
say, "Hang on a sec. That nerd might own his own submarine in about
10 years."
Rob Collie (00:12:21):
With or without missiles.
Hugh Millen (00:12:22):
Yeah, right.
Rob Collie (00:12:23):
Yeah, nerds are having a little bit of a moment. I'll give you
that. I don't really think that the star athlete has really lost
too much luster.
Hugh Millen (00:12:32):
Yeah.
Rob Collie (00:12:32):
I think still ... they both retain it. So this is a question I've
never actually asked you. When was the first time you discovered
your interest in data? What's your data origin story?
Hugh Millen (00:12:44):
Well, I think there was a couple of inflection points probably in
that regard. When I was going through a contract negotiation, this
would have been 1992 and I was collecting data, I was doing some of
the legwork for my agent, Marvin Demoff was my agent at the time
who had Dan Marino and John Elway, and he had a lot of high profile
athletes, far more accomplished than me but at the time, we were
trying to extract the highest contract in the history of the
Patriots. So we had to kind of compare all the quarterbacks, the
starting quarterbacks and this was back in Lotus 1-2-3.
Rob Collie (00:13:21):
That's okay. We had Mr. Excel on a few episodes back and even he
started in Lotus 1-2-3.
Hugh Millen (00:13:22):
Yeah.
Rob Collie (00:13:27):
It's okay, we're beginning.
Hugh Millen (00:13:28):
Right, so it was lotus and ran a couple spreadsheets and a couple
rudimentary formulas, and was a friend of mine and he was able to
kind of manipulate this data and I kind of slotted myself at about
the top of the fourth quadrant of quarterbacks. I mean, I wasn't
trying to say, "Hey, look, give me Jim Kelly, Dan Marino money." I
was down with the slap ease, but I thought it was pretty cool to be
able to that on a home computer, right? I had a home computer,
doing other things but I hadn't really got into data. So that would
be one area where it kind of planted a seed like, "Hey, you need to
acquaint yourself with the powers of a spreadsheet." Well, then
once I got into Excel and started to discover its power, it was
just so fascinating to me. I would say another point worth
mentioning was, I had a job, it was really an avocation, discussing
the University of Washington, the Huskies football team on the
radio.
Hugh Millen (00:14:25):
There was a point where there was some day where, for whatever
reason, as I kind of log the plays by hand, each play, of course,
would be a record, right? I was logging the formation and the type
of play and the defense and I made a comment, I'm going to butcher
the exact details but it's pretty close, I'm going to get it pretty
close. For whatever reason, I said, "When the Huskies were on the
left hash ..." and this is me on the radio, when the Huskies are on
the left hash and they threw an in breaking route to the widest
receiver and so, there may have been another variable, but let's
just say for the discussion, there was at least three variables. I
said those three variables had to be precedent. I said, then the
Husky quarterback, his name was Cody Pickett, at the time. I said
Pickett had some astronomical success rate. He was ... I said
something like that. He was nine for 11 for 175 yards and what have
you.
Hugh Millen (00:15:24):
I said, but on all other passes, he was only 12 for 26 or something
for 120. So I was able to, mentally just hit me that when those
conditions were present, he have had this success, and then it
might have been that day of practice or no later than next day. The
quarterback coach from the Huskies, comes walking over to me and
says, "Hey, tell me that stat you said on the radio about ..." I
recited what ... so he was unaware of it and I had just kind of
stumbled upon it by chance and I just realized that there's an
interest there and at least some domains about the specificity of
the data, that the data, it's laying out there before us and maybe
to an observer ... maybe it would be like, staring at a chessboard,
where you'd say, "Wait a minute, if you just make one move with
your knight and another move with a bishop, you got checkmate." You
could stare at that chessboard, and you don't see it.
Hugh Millen (00:16:22):
So I think data could be like that, where there's something that's
really telling and meaningful to people but unless you have a means
of crunching the data and extracting the data, then you might be
oblivious to it, like somebody staring at a chessboard unaware of
how close you are to success.
Rob Collie (00:16:44):
Yeah, there's two things in there that I want to call out. First of
all, something that we say all the time in the business world, is
that sooner or later, every individual is going to have a
collision, a professional collision with the spreadsheet and at
least 15 out of 16 people bounce off of that spreadsheet. You get
the hell away from that thing but about one out of 16 at most, one
out of 16 human beings, when they collide with the spreadsheet,
they stick?
Hugh Millen (00:17:17):
Yeah.
Rob Collie (00:17:17):
You never know. We call it the data gene. The data gene is what
determines whether you stick or bounce.
Hugh Millen (00:17:22):
Really.
Rob Collie (00:17:23):
If you have the data gene, you stick. Data gene is rare. It's not
super rare but it's 5% or less'ish of the population and it cuts
across every single demographic.
Hugh Millen (00:17:35):
Yeah.
Rob Collie (00:17:35):
I love that. I can't believe that I've never asked you when that
collision happened for you, and it was in a 1992 contract
negotiation. How fascinating.
Hugh Millen (00:17:45):
Yeah, right and there's been other things. As I learned it, I had
some people that started to rely on me at the station. I'm just
like, "Hey, we're running a contest, and we want to have people be
able to predict college football games," but they can only do it
once per week and they can't use the same team. So now we're
getting 10,000 people that are texting in and there's this poor guy
that was doing everything manually.
Rob Collie (00:18:14):
That sounds like Luke.
Hugh Millen (00:18:17):
He was just manually sorting and he was spending six hours a week,
I'm not kidding.
Rob Collie (00:18:22):
At least.
Hugh Millen (00:18:23):
Yeah, right and I just said, I could take an hour and a half and
write you an Excel program or Excel worksheet. program.
Rob Collie (00:18:30):
Yeah, a program. A program is a good word for it.
Hugh Millen (00:18:32):
A workbook, I guess would be the technical because it had several
worksheets that all referenced each other.
Rob Collie (00:18:38):
Yeah.
Hugh Millen (00:18:38):
So there was about 10, 12 worksheets within the workbook. If you
want to call it program, fine but I guess that would be the most
technical way based on my own knowledge of the vernacular.
Rob Collie (00:18:49):
You're correct. Worksheet is the official term. One of the things
we always try to highlight on this show is that a spreadsheet
itself is an application, we think of Excel as an application but
when you create a workbook, a worksheet, whatever, a spreadsheet,
you're programming. Even formulas are a form of programming and
that thing that you produce, you basically ... you built this guy
an app.
Hugh Millen (00:19:12):
Yeah, well, whatever we call it, he was grateful to go from four
hours, four to five hours a week down to about 10 minutes. They had
to load the data. So I had set out a portion of a week worksheet
that he could just dump the raw data and then all the other
worksheets reference that. Actually, there was another one with
golf. I did at least two or three applications for the radio
station and I think Steve Balmer at one point even referenced that
in one of his discussions. He was saying ... giving two or three
examples of how Microsoft products help in all kinds of strange
ways. He says, "Hey, this former NFL quarterback is writing
spreadsheets to help a radio show execute their radio contests." I
don't know, it's here I say I heard that he had said that.
Rob Collie (00:20:02):
It sounds consistent with the Balmer I know of.
Hugh Millen (00:20:04):
Basically, I've said it many times. I think Excel is the coolest
spreadsheet on the planet.
Thomas LaRock (00:20:09):
I was going to say a few things. Hugh, wonderful to meet you.
Hugh Millen (00:20:12):
Yes, likewise. Thank you.
Thomas LaRock (00:20:13):
So glad you and Rob cross paths, so now you and I can cross paths.
Although one thing I want to say is, as Patriots fan for many
years.
Hugh Millen (00:20:21):
Yeah.
Thomas LaRock (00:20:21):
People always question that. They're like, "Oh, you just like in
this ..." I go, "No, no. No, no." I remember Rod Rust. I remember
Hugh Millen.
Hugh Millen (00:20:31):
Okay. Yeah.
Thomas LaRock (00:20:31):
When we logged in, you saw I had Tommy Hudson.
Hugh Millen (00:20:33):
Yeah. Yeah.
Thomas LaRock (00:20:34):
This is my youth, Okay?
Hugh Millen (00:20:36):
Yeah. I got it. Yeah.
Thomas LaRock (00:20:36):
Right?
Hugh Millen (00:20:37):
Sure.
Thomas LaRock (00:20:38):
So, it's an honor to be able to have this conversation with
you.
Hugh Millen (00:20:41):
Likewise. Thank you.
Thomas LaRock (00:20:41):
Then, I do a little research and I find out you're a Husky and I'm
like, "Well, everybody has to go school somewhere."
Hugh Millen (00:20:48):
Yeah.
Thomas LaRock (00:20:49):
I did, my graduate school was Washington state.
Hugh Millen (00:20:53):
Okay, you're a Coug. Okay and where was your undergrad?
Thomas LaRock (00:20:56):
Merrimack College, north of Boston.
Hugh Millen (00:20:58):
Okay.
Thomas LaRock (00:20:58):
When you mentioned Eastern Washington. I'm like, "Oh, yeah, Cheney.
I know where that is"
Hugh Millen (00:21:02):
Cheney, yeah.
Thomas LaRock (00:21:02):
No, problem, right?
Hugh Millen (00:21:03):
Yeah.
Thomas LaRock (00:21:03):
Even having this conversation, I'm kind of getting some nostalgia.
I'm like, "Ah, I remember all these places and things, but here's
the thing I want to get to, is the stats, the level of let's say
data in the NFL at that time because you just said, "Hey, I knew I
could do this thing in Lotus 1-2-3 in order to leverage it for a
contract negotiation," and I have this picture of you walking up to
Billy Sullivan, with a spreadsheet and you show it to him and he's
probably looking at it and go, what the hell is this thing?
Hugh Millen (00:21:37):
Yeah.
Thomas LaRock (00:21:37):
What does it mean? I'm just wondering, so my first question, did
you get the contract, and my second question is what was their
reaction when you were using data to make the case at that
time?
Hugh Millen (00:21:50):
I made the notebook and pass it on to the agent, because he was
going to be doing the negotiation. So to me, I was obviously acting
on my own behalf to try and help me. So, I don't know what their
reaction was other than second hand and what my agent said is that
we made a reasonable case out of it. So I ended up ... yeah, I get
the highest contract in the history of the Patriots franchise and
they had been around for 33 years. So I think on some level, we'll
never know the variables, you can't isolate the variables. I don't
know what contract I would have got without that, but presenting
them some dat, and again, it wasn't like I was at the top of the
league. I was, as I said, probably at about the 75th percentile, I
was trying to clock in. So I was trying to be better. There's 28
teams, I was probably trying to be in the top ... barely the top
20. That's probably what I was trying to do. Yeah, it ended up
working.
Thomas LaRock (00:22:48):
So I often say this, I don't mean to be disparaging in any way, but
I tried to tell people I've been a fan for the Patriots for a long
time and when we talk about that era, I remember the season. You
guys went six and 10. I call into Ordway Show and I'm like, "Hey,
patriots, they could have won 10 games. They should have at least
had nine. They should be in the playoffs in Ordway." He was polite,
I guess back then but he was like, "No, they're not that good." So
the following year, there's a bit of a dip.
Hugh Millen (00:23:18):
Yeah.
Thomas LaRock (00:23:19):
I think you were hurt a little bit as well and of course, Dick
MacPherson. That's always a factor. There's the Dick MacPherson
factor, but I tell people, without Hugh there is no Drew. Because
that one season you had where there was the glimmer of hope that we
had the town and the ability to do something when the Sullivans
then had to start thinking about selling and everything new and
then Parcells comes in. We hit that low spot with Rod and Dick and
Parcells comes in and the first thing we do is get a chance to
draft somebody to be a potential to be a top level, somebody who's
just going to sell tickets. It was a reboot of an entire franchise
after 35 years, and it was a fabulous ride.
Hugh Millen (00:24:03):
The 1990 team, that team had gone one and 15 and I was on the
Falcons and played the last couple games, so I had some contract
offers, I got nine contract offers, but every one of them, other
than the Chargers where they said, hey, it's an open competition
with three quarterbacks and the Patriots, it was an open
competition with two, just me and Tommy Hodson. So they were giving
me the best chance but they had been one and 15, the year before
and in fact, the NFL Network, they do their top 10 best rivalries,
top 10 rookie running backs. They do all these top 10 shows and
they did the top 10 worst teams. So obviously, it's naturally the
bottom 10 but the top 10 worst teams and the Patriots of 1990 were
deemed by the NFL network to be the eighth worst team in the
history of football.
Hugh Millen (00:24:51):
I mean, like there's like the Canton Bulldogs and all this stuff. I
mean, we're talking about 100 years, Bill Belichick had offered me
more money to go to the Browns than the Patriots had and he was the
head coach of the Browns at the time, but I wasn't given an
opportunity to compete for the starting position. So at one point,
I was thinking between the Chargers they had, as I said, a three
way competition, Patriots two. Dick MacPherson, as you said and
guys, he was a legendary college coach at Syracuse, so he was used
to recruiting. I was in the College Hall of Fame at some point. So
now he's talking to me. He's trying to recruit me to a one and 15
football team from the year before. He knows that I'm considering
some other teams, including the chargers, and I remember him
saying, "Okay, you tell me why the Chargers are better for you than
the Patriots and don't say the beach."
Hugh Millen (00:25:46):
He was right, that was my best chance. So for us to go six and 10
they were writing articles. Kevin Maddox for example, you know that
name. I mean, he was writing ... That was the most exciting season
in Foxborough history, just like all of our games were close. We
beat the Bills, they were 11 and one and we lost some other games
right down to the last minute. I think we had like four 4th quarter
wins out of the six. So it was based on having been won in 15, the
six and 10 was good and then the next year, yeah, I hurt myself in
the first quarter of the first game. Seventh play of the season.
Separated my shoulder. We went two and 14 and then everybody got it
out of there. As you said, that was Parcells, he came in and as did
Bledsoe and I was out.
Thomas LaRock (00:26:27):
So, also, all the things that we shared, not just the Washington
History and Patriots but Dick MacPherson. I just laughed. I'm like,
"That's right. He played for Dick MacPherson." You mentioned him as
a recruiter. Dick MacPherson came to my high school, I was playing
a basketball game, he was there to recruit, a visiting player by
the name of Mark Chimera and he sat in the bleachers next to my
grandfather, who they chatted each other up for the entire game. I
just kept looking at the stands like, what are they talking to each
other about? Who do they possibly have in common? Yeah, Dick
MacPherson in my gym and I always laugh about it to this day and I
think he was ... obviously he was still with Syracuse, but he was
on his way out the door. He was trying to get mark before he went
to Boston College, but he was on his way out the door heading to
the Patriots, I think already.
Hugh Millen (00:27:16):
Well, I can say with sincerity, if you ask me, okay, the most
beloved person I've ever known and it's okay, well, what does
beloved mean. The highest number of people that could respond, that
they view Dick MacPherson or anybody so affectionately, that they
could describe it as they ... without knowing the person, they feel
like they love the person, find some reasonable definition of
beloved and it's quite possible that he's the most beloved person
I've ever known. Now, Don James for whom I played in University of
Washington, he's in the Hall of Fame as well and he was immensely
respected, but he wasn't beloved like Dick MacPherson. I have great
affection for the man, Dick MacPherson. I can't imagine anybody who
was ever sideways with Dick MacPherson. Not much to do with data
guys, but that's a trip down memory lane. Yeah,
Rob Collie (00:28:07):
I did want to get this in relatively early. If people who are
listening, they're like, "Ah, are they turning our data podcast
into sports talk retrospective?" No, it's certainly spicy, right?
It's very interesting and we're going to season the conversation
appropriately. One of the things that I want to bring front and
center very early is, why are we so fascinated, why does sports
analysis ... when it comes to data, why does sports come up so
often and why is it relevant to talk about sports data and sports
analytics, to a business person? Not everyone that listens to this
podcast is a sports fan, but everyone listens to this podcast as a
data fan in some form or another, right, in a professional sense.
Why is it and I believe that it is, I believe that talking about
sports analytics is actually an incredibly powerful sort of
learning tool. Even if you're never going to perform sports
analytics, if all you're going to be doing is business
analytics.
Rob Collie (00:29:10):
First of all, the data is public. The data is public. The data is
shared. There aren't many data sources that you can crunch on, that
are both interesting and have relatively high stakes. There's a lot
of money on the line, at least for the organizations involved.
There's a lot of pressure, a lot of pressure to succeed, and data
is public. Every last play of an NFL game or any football game for
that matter is now recorded from 18 different camera angles is
dissected every which way, like there's nothing ... on the surface
anyway, there's nothing hidden. All the data is out there. It's
like a freaking public domain. We don't have too many things like
that. The number of COVID dashboards that have sprung up in our
community over the last year. I mean, it's like there's almost more
COVID dashboards than there are COVID cases.
Rob Collie (00:30:02):
I think it's the same thing, right? It's a public dataset,
relatively public anyway, with consequences. If you've ever been
interested in the COVID dashboard, you might as well be interested
in the sports dashboard, even if you're not into sports, like I'm
not into catching viruses either.
Hugh Millen (00:30:16):
Yeah. Imagine a sports world without statistics. Sports world
without statistics, you never knew that Joe DiMaggio had a 56 game
hitting spree. How do you even compare the completion percentage of
the passer rating? What's Michael Jordan's scoring average? It's
almost like you wouldn't even enjoy the entire world of sports. You
referenced music earlier. Okay, I don't think you need statistics
to enjoy music. If you like The Rolling Stones, now you might
tabulate how many albums they have, or the Beatles or something or
songs that made number one, but I don't think that that's central
to the enjoyment of it, but I almost feel like, if you didn't have
statistics, it would be far less interesting if you weren't aware
of how the teams and how the players are doing and to that end,
after an NFL game, for example, they have what's called an NFL game
summary. They have very capable statisticians that are logging
these plays.
Hugh Millen (00:31:19):
Obviously, there's an application there that's sorting out the
statistics, within 12 minutes of the game being over, they have a
17 page game summary of all statistics.
Rob Collie (00:31:32):
Yeah. It's crazy.
Hugh Millen (00:31:33):
Almost an on ... play by summary there, but of the 17 pages on
average, there's 12 or 13 of them are stats. So yeah, you're right,
it's difficult to conceive of sports without statistics.
Rob Collie (00:31:47):
Well, when you think about it, and I only think about it now
because of what you said, even who wins is a question of
statistics. It's a number on a scoreboard.
Hugh Millen (00:31:57):
Right.
Rob Collie (00:31:57):
It determines who wins, right? Someone, somewhere decided that a
field goal is worth three points and touchdowns is worth six and
all of this, right? Those are coefficients. It's a it's a weighted
average of your performance, right? If we decided that field goals
were worth five, it changes the whole game.
Hugh Millen (00:32:12):
Yeah, yeah. Excellent point. I just think that being able to
comment for me to comment on the radio, to find reflection, for
example, we're out here in Seattle where Russell Wilson is an all
pro level quarterback for the Seattle Seahawks. Most people think
he's bound for the Hall of Fame. He had a really tough year in the
second half of the year. So just yesterday, I was on the radio, so
the end game is ... and this is kind of typical how I use data. I
say, okay, I start with a question that can be asked in plain
English, without any statistic. I want to know where the Seattle
Seahawks is more aggressive in terms of their propensity to throw
the football, were they more aggressive in the first half of the
season, versus the second half of the season. Okay now, how do I
determine a reasonable way to measure that?
Hugh Millen (00:33:07):
Well, it's easy to just take the entire data of all plays and say,
"Okay, this is the run percentages. This is their past percentage,
right?" Because every play is either a run or a pass so we have
that aspect, but a lot of times teams when they're ahead, the
better teams, they're trying to kill the clock, they're more likely
to run or teams at the end of the second quarter, if you happen to
have a drive at the end of the second quarter, you might have nine
straight pass plays. Well, that can skew things. So there's these
elements, so it was pretty crude but what I said was okay, I'm
going to go the first and third quarters, because that eliminates
the second and fourth quarter, the second quarter can have those
two minutes situations that I described. Fourth quarter, if you're
behind, you're going to throw it all the time, if you're ahead,
you're far more likely to run.
Hugh Millen (00:33:53):
So I said, first and third quarter. Now, I also wanted to eliminate
third down because third down is a down that, if it's third and
long, you're more likely to throw it. If it's third in the yard,
you're more likely to run it, whatever. So I want to eliminate
third down. So really the neutral down, so I said let's go first
and second down and then, let's go yards to gain between two and 10
because if I do that, I'm picking up all first and 10s and I'm
picking up all second and two to 10s, so I'm eliminating second and
15s when you had a sack or penalty or something. So, it was a
roughly crude way to do it. Well, as it turns out the Seahawks in
the first half of the season, they were second in the NFL in terms
of their pass propensity, pass-run ratio, Second only to the Kansas
City Chiefs who were the world champions at the time. They're not
the world champions now, but you get my point.
Rob Collie (00:34:45):
Yup.
Hugh Millen (00:34:47):
In the second half of this season, they fell to 18. So that's
something that's meaningful enough that I can take that to the
radio. It's just simple crunches. It's under three, four minutes.
Three minutes, I can ascertain all this and then I can decide is
that something that's compelling enough to present to a radio
audience? Well, in my estimation, it was.
Rob Collie (00:35:10):
It sounds good to me. Yeah.
Hugh Millen (00:35:11):
Yeah, if you lay out a reasonable standard that would define what
you're searching for, because when numbers interact, you get these
rankings, right? You're going to be second place and 18th. Had it
been ninth and 11th, I wouldn't have taken it to the radio but the
numbers ... so you've got to think through, okay, feed these
numbers that are binary in the equation and then try and glean some
data that's meaningful.
Rob Collie (00:35:36):
So if you're listening, and you're not into football, let me kind
of break that down for you a little bit. You just heard a human
being who used to get paid millions of dollars to play football, go
through an analysis where he compensates for and filters out all
the external variables that would confuse the hypothesis, right,
and dial out the situations. So basically, like we could sort of
talk about ... you got down to the situations, you filter down to
the situations where the Seahawks faced neutral situations.
Situations where the external variables don't indicate strongly
that they should do one thing or another.
Hugh Millen (00:36:20):
Correct.
Rob Collie (00:36:21):
In order to find out what they did when they sort of had a
choice.
Hugh Millen (00:36:24):
Correct. That's exactly right. That's better stated, yup.
Rob Collie (00:36:27):
We call this ... for lack of a better term, we call this fair
metrics at our company. The top level metrics for a business or a
sports team, they sort of like appear at the bottom line, are so
conflated, they're so confused with so many other external
variables, that you can't use them to make decisions most of the
time, right? You've got to do something like what you just did.
You've got to develop the more fair metric, right? There are
versions of this running around, I'm sure but we could try to
trademark one now and call it like neutral situation tendency rank,
dropped from second to 18th in terms of their decision making.
Hugh Millen (00:37:04):
Yeah. Now, if I'm going to confess and this is something that I'm
assuming, if you're listening to this podcast, you have a keen
interest in data. Now, I'm going to have to bring up a little topic
that all of us who work in data have to confront, which is I also
ran the data with all the variables I just described, and I ran it
when the score was plus or minus eight points, meaning within a
touchdown either way, because of touchdown two point
conversion.
Rob Collie (00:37:35):
Yup.
Hugh Millen (00:37:36):
The data there was the Seahawks were number one in the first half
of the season, but only fell to like, I don't have it in front of
me, it's like number seven. So now, it doesn't sound like as
drastic of a drop. Yeah, there's almost a moral dilemma to me
because we all know that we can find ... if we're looking for a
conclusion, we can scour the data enough and we can lock onto the
one and say, "Aha, that's the one," now it's factual.
Rob Collie (00:38:11):
It is.
Hugh Millen (00:38:12):
I would never use false information, but is there an obligation we
have morally to disclose the fact that hey ... and in fact, I will
do that. I will say ... from time to time, I will just say, "Look,
I thought I had this hypothesis, I ran the numbers and actually
didn't pair out," and I will confess to that but there's other
times where I kind of like, "Hey, I run the numbers. Yeah, it
didn't really support my ..." and I'll rerun them in some other way
that changes it. "Okay, I like that number. That's going to sound
better on the radio." I would imagine that Rob, certainly, you've
encountered that, where you ran the number, it didn't come up to
the results you wanted and let me try and run it a different way,
right?
Rob Collie (00:38:59):
Back in the day, we've even been fired for giving the conclusion to
the client that they didn't want. It's like, we don't really have
those kinds of clients anymore but nearly going, we did have a
couple it, apparently hired us just to help support their original
hypothesis but never told us that.
Hugh Millen (00:39:17):
Yeah.
Rob Collie (00:39:17):
Let's keep going on that exact example you talked about, I think
this is both very interesting and also it reflects the integrity
and the curiosity that I think you really need to bring to anything
like this. So when you become more specific, and you're dialing out
... so what you're saying is like, in one analysis, I removed ... I
controlled for essentially, some number of external factors and in
that analysis, they dropped from second to 18th. Okay, but then I
added an additional external factor that I controlled for, so I
became even more specific and the Delta isn't as large in terms of
their ranking drop there. Now, of course, at the same time, you've
also reduced your sample size by filtering down further, you're now
looking get fewer and fewer plays and I think it's in the book
Fooled By Randomness that talks about this college professor in
statistics, who starts every semester by challenging the entire
class to like a duel.
Rob Collie (00:40:13):
It says okay, "We're going to randomly ..." and I won't know which
50% is which 50%, "We're going to randomly assign half the class to
flip a coin 30 times and record the results factually," and the
other half of the class gets to or has to fake it. They've got to
write down 30 coin flips, as if it was fake.
Hugh Millen (00:40:39):
I know how you differentiate. Yeah.
Rob Collie (00:40:41):
Yeah and the class knows, obviously, who was who but he doesn't.
Then, they all turn them in and he just very confidently goes
through them, and puts them in the, "You were real, you were fake.
You were real, you were fake," and just nails it overwhelmingly.
The reason he can do this is because in reality, there's always
going to be some really improbable sequence of consecutive heads or
tails. It's going to happen, you're going to get five in a row, and
a human being would never believe that, right? So when they're
doing it, if they've got three or four in a row, if they've got
even three in a row, they pretty quickly flip back to the other
one, right?
Hugh Millen (00:41:19):
Yeah.
Rob Collie (00:41:20):
The point there is, is that as you get to smaller and smaller
sample sizes, there's going to be something compelling. That jumps
off the page or potentially anyway. You can't trust the small
numbers, right? So, it might be that your theory ... I agree that
like falling from second to 18th, that's a compelling narrative.
First to seventh isn't so much, but at the same time ...
Hugh Millen (00:41:45):
If there's 32 teams, yeah.
Rob Collie (00:41:47):
It might be, that's a good point, right? People don't really
necessarily know how many teams are there.
Hugh Millen (00:41:49):
Yeah, if there's like 500 teams then.
Rob Collie (00:41:52):
Yeah, it's definitely not. It might be that what you're discovering
was still holding up really well. It's just that with a smaller
sample size, you got the Fooled by Randomness thing sort of
compensated the other way. Furthermore, there might be a huge
difference between first and seventh. What if teams seven through
32 are all clustered like together in one clump of tendencies?
Hugh Millen (00:42:15):
Yes.
Rob Collie (00:42:15):
And there's a really sharp slope from one to six. It could still be
a huge difference, right?
Hugh Millen (00:42:22):
Yeah.
Rob Collie (00:42:23):
Look at this, even this one thing you're talking about is so
fascinating, like you can just like really just dive into it.
Hugh Millen (00:42:31):
Yeah, but then you got to be knowledgeable of your audience.
Rob Collie (00:42:34):
Yeah.
Hugh Millen (00:42:35):
And know that they can handle some statistics but you can't get too
dry. It's got to be something that, for me to start citing
statistics, it's got to be comprehensible and it's got to tell a
real meaningful story. I've got to feel like I've accomplished
something by presenting data.
Rob Collie (00:42:57):
Yeah.
Hugh Millen (00:42:58):
It's usually you're buttressing your argument, you mentioned that
you ... the ever present availability of the statistics, if you
take some quarterback, let's say whether it's Ryan Fitzpatrick or
Josh Allen or Deshaun, Watson and I've just mentioned a guy who's
close to average, and then two guys that are pretty damn good but
not at the almost at the top of the NFL. I'm pretty confident that
if I wanted to pick one of those quarterbacks, and I really wanted
to steer it towards a narrative, I could come up with enough data
and really compelling sounding data that would make those guys
either ... for example, Josh Allen, quarterback of the Bills, I
could probably come up and say, make an argument, he's as good as
anybody in the NFL or I could probably find data, if I wanted to
push the narrative that he's just average. He's dead middle. I
could probably find data. Now, most people, would just say, "Hey,
look, I've watched the guy play."
Hugh Millen (00:44:02):
"He's ... for this most recent years, he's in the top five," but
you'd be surprised how compelling you could present a statistical
encapsulation of guys and so data can be really powerful.
Rob Collie (00:44:16):
Yeah, I think NFL GMs are wise to that today, whether they're data
savvy or not, they know that the data can be used to paint one
story or the other, so they know that it's sort discounted if it
comes from a slanted source, a potentially biased source, but in
1992 they were defenseless.
Hugh Millen (00:44:33):
Yes. It's like, what is this? What?
Rob Collie (00:44:40):
Give them the money.
Hugh Millen (00:44:42):
You know what's funny on the front page ... so the team never
supposed to know that I did it, right? I took some clip art.
Remember the clip art? Remember clip art? Remember now, this is
1992. So we got some clip art of like a referee holding his hands
up like touchdown, like some subliminal message like, "Hey, as a
sign of humility, you're going to get more touchdowns," which the
statistics actually didn't bear that out but maybe the clip art
would, right? I remember the agent kind of like ... he goes, "I
don't know about the clip art," and he goes, "Ah, we'll just leave
it in." It's like he had bigger fish to fry than me and he had
bigger fish to fry than trying to take out the referee in his
raised arms but 27 years old and you're new to computers or ...
computer is in evolving stage, you do silly stuff like that,
right?
Rob Collie (00:45:26):
They were defenseless. We're known some circles as the stick figure
people of data. We use a lot of clip art. Tasteful, tasteful clip
art
Hugh Millen (00:45:36):
Yeah, right. Good action. Well, yeah, we've come a long way.
Rob Collie (00:45:42):
When you're talking about, you've got to be careful to still be
compelling to your audience, there's a parallel there in the
business world as well. Think of it this way, you're one of the 5%
of sports people, fans or industry figures, whatever 5% of people
who are interested in sports, who have the data gene. So when you
get on the air, you've got to be cognizant of that 95% that only
have a certain amount of attention span for it, they're not as
interested in it as you are and you can't make the mistake of
thinking that everyone else is as interested in this stuff and is
willing to kind of like Rainman it with you.
Hugh Millen (00:46:20):
Sure.
Rob Collie (00:46:20):
That you will and that's true in business. The 5%, people who are
running around in business with the data gene, aren't typically
running things. So not only are they outnumbered like 19 to one in
their organizations, the people who outrank them tend to be in the
19. Even more than statistics would indicate.
Hugh Millen (00:46:42):
Sure.
Rob Collie (00:46:43):
So, we have to be relevant. We have to be digestible. We have to be
actionable. All those things you talk about, that you go into when
you're performing your analysis before you slide up to the
microphone, that is 100% true in the business world as well and
like, there are many, many, many stories of heck, even my own
personal experience with this stuff, where I kind of overestimated
how interested the audience was, and ended up kind of alienating
them accidentally in the process, and therefore losing our
opportunity to find something meaningful because like, they kind of
became less interested in data as a result of me being a little
tone deaf. I've learned a lot over the years, I don't think I'm
nearly as likely to repeat that mistake, as I was like in the
earliest days of this, but it's a road all of us have to walk.
Hugh Millen (00:47:35):
So what we don't want to do is try and impress the audience, by the
fact, "Hey, look, I know how to manipulate data and I can write a
spreadsheet with formulas that are three lines long." Well, we want
to say what's the end game? The end game is the data should be
illuminating and it should be like, the numbers supporting an
argument that you can speak with words. Does that make sense? It's
like, think of the SAT. You've got the verbal portion and the math
portion, right? So in that analogy, we're making arguments that can
be structured merely in words and then we bring in the math, and we
try and make it as simple as possible to then augment the argument
that we're making with the words only. Does that make sense? So, I
could talk real fast on the radio and I could start seeing data and
use some of the vernacular within Excel and confuse the hell out of
people, right?
Hugh Millen (00:48:37):
Some people say, "Oh, boy, he sounds really smart. He lost me."
That's not the end game. I am willing to have people say that I'm
hard to understand in terms of the schemes that I'm presenting but
I try all the time to really dumb it down. For example, in
football, there's a defense called cover three. It's a three deep
zone, that may or may not mean much to you. Then, what I'll do is
say, think of it like this, think if you got four defensive lineman
and rushing the quarterback. Okay, now we're down to seven guys.
Now, think of like a baseball outfield, you've got the infielders,
you got the third baseman, the shortstop, the second baseman the
first baseman, because I know people in their minds eye. If they're
listening to sports radio, they can do that. You can imagine okay,
you're hovering over home plate and you're looking out at the
baseball defense and then you've got the left field or the center
field and right field.
Hugh Millen (00:49:31):
I said, a 3D zone defense is really similar to that. You got four
guys underneath. They're like your infielder, they're the guys
closest to the line scrimmage and then the three guys that are in
the deep layer, they're like the outfield. Where's the
vulnerability in baseball? Where do you get the big hits? Well, you
hit him in the gap between the left fielder and the centerfielder
and we have past routes that can kind of hit that area. We call
them skinny post and deep in routes. Then there's also, you can hit
it double down the line, okay, to the outside of the outfielders.
We call those corner routes and comebacks and what have and those
are the sidelines. So I'll take the time to stop and I'll describe
something I'd figure that most people don't know about.
Rob Collie (00:50:13):
I love that.
Hugh Millen (00:50:14):
I'll try and describe it in a way where it's a cover too. Imagine
if you were a soccer goalie, you're going to do what? You're going
to just stand in the middle of the goalpost, right? Now imagine
just for a second that the rules of soccer allowed us to have two
goalies? Let's say Thomas, you and I are allowed to be the goalies.
All right.
Rob Collie (00:50:35):
Can I jump in for a moment?
Hugh Millen (00:50:36):
Yeah, please.
Rob Collie (00:50:37):
If I come up on a soccer goal, I'm supposed to score and I see Hugh
Millen and Tom standing there, I'm going to kick it at Tom every
time.
Hugh Millen (00:50:43):
Yeah, go for Tom. Yeah, go for Tom. What's up with that Tom? You
didn't deserve that. Okay, I'll say imagine in soccer, if they said
for, let's say half of a game, that they're going to allow two
goalies. Well, how would you and, Tom would stand. Well, logic
would dictate that we'd stand in a way, we kind of dissect each
half, right? So that if Rob had some killer thunder foot shot,
right down the middle, right between us, he could beat us that way,
or a killer thunder foot shot right next to the goal post. That's
how we're going to space ourselves so that's where our
vulnerability is going to be. Well, that's like a cover two
defense. You got a safety on each half of the field and there is a
vulnerability right down the middle and there's a vulnerability all
the way deep on the sidelines.
Hugh Millen (00:51:34):
So, I'll constantly try and think of ways to describe things and
then, use the data with ... in my mind, a lot of discernment
whether it's comprehensible because I also have to think of this,
I'm talking to people who are driving to their meeting or driving
home and they're thinking about their day. They're not sitting in
earnest, listening at the desk with nothing else going on, perhaps.
So I got to take all of that into account, but getting back to the
data portion of it, there's kind of a barometer that I run through
on whether I'm going to present the data. Is it comprehensible and
does it fulfill some objective of I'm trying to get from point A to
point B. I'm trying to be persuasive. Usually, I'm trying to be
persuasive. I'm giving you my take, and I'm trying to get you to
agree with my take and I'm trying to give a strong take. So, if I'm
introducing data, it's in an effort to have a stronger take, it's
really what it is
Rob Collie (00:52:35):
That thing were used to sort of dumb down, what cover three is.
Hugh Millen (00:52:39):
Yeah.
Rob Collie (00:52:40):
I think that's it's just like, "Ah, that's doing God's work, in my
opinion." Dumbing things down and getting through all the jargon is
so valuable in any domain. I was just sitting here, I was listening
to you, going like, "Oh my God, we should do a podcast called like
football dumb down" or something like that.
Hugh Millen (00:52:59):
Yeah, yeah.
Rob Collie (00:53:00):
If you heard the story, it's actually an interview, where he's
speaking at a dinner or something. Brett Favre talk about the
Nickel Defense with Holmgren
Hugh Millen (00:53:08):
Yeah.
Rob Collie (00:53:09):
Have you heard that?
Hugh Millen (00:53:10):
Yeah.
Rob Collie (00:53:10):
He, Brett Favre was like three years into his NFL career and didn't
know what a nickel defense was. I mean, it doesn't matter whether
you know what a nickel defense is or not, it doesn't matter. Brett
Favre didn't know, right? When he finally got, what was it like Ty
Detmer to explain it to him.
Hugh Millen (00:53:28):
Yeah. Yeah, you got it.
Rob Collie (00:53:29):
Ty Detmer said the nickel defense is when like they take out a
linebacker and put in another defensive back. Favre goes, "That's
it? Who gives a shit?"
Hugh Millen (00:53:38):
Yeah, that's it. You got it exactly right. Yeah, so he make ... so
they set the standards for DBs, you bring in one more guy defensive
back that is and it makes the fifth, hence nickel, five, right?
Rob Collie (00:53:51):
It's like the old Chris Rock joke when he's saying like, "You can't
even tell your kids anymore, not to smoke crack because the mayor
of DC has smoked crack." He's like, "You can't smoke crack. What do
you want to be when you grew up?" I could be mayor. How do you
expect to succeed in football if you don't know the difference
between a nickel defense and a standard defense? Well, I could be
Brett Favre.
Hugh Millen (00:54:17):
Yeah. Well, Favre was an outlier in that regard, right? Let's
underscore that point. He had so much talent that they could have
put 15 guys out there and he still would have found a way to figure
it out, that's a good option.
Rob Collie (00:54:37):
It's so awesome. Well, here's the thing, NFL, whether you're
interested in it or not, it is a multi, multi, multi billion dollar
industry. It's huge money. I think I'd rather live in a world where
the NFL was less important. Even though I love the NFL, I love
football and everything. I think I'd prefer a world whose
priorities were a little bit more aligned with overall human
thriving.
Hugh Millen (00:55:03):
Yeah.
Rob Collie (00:55:03):
I don't get to decide. So in the meantime, it is huge business. You
can't really even think of a place where better decision making in
terms of like team construction or team strategy, you can't really
think of a place where there's more money at stake for success.
It's not just for our entertainment, successful teams are really
good business.
Hugh Millen (00:55:27):
Yeah.
Rob Collie (00:55:28):
This is something you and I have been ... you've been fascinated by
this longer than I have, but that same industry that we just talked
about that has tremendous resources and everything in the world
possibly at stake, we just watched Tom Brady at age of like, is he
like 60 yet. We just watched him win yet another Super Bowl with
yet another team.
Thomas LaRock (00:55:47):
He is 43.
Rob Collie (00:55:48):
He's 43. Okay. I was just rounding him up to 60, and we're like one
Drew Bledsoe injury away from having potentially never even known
that he existed.
Hugh Millen (00:56:01):
Yeah.
Rob Collie (00:56:01):
This really highlights, despite all of the resources, despite the
rise of analytics, and its popularity and how front and center it
is now, I still don't think they've even begun to figure it
out.
Hugh Millen (00:56:13):
No, they haven't, and that's borne out by how much they fail in
assessing quarterbacks. As you mentioned, Brady, we all know he was
a sixth round pick, 199th overall pick in the draft. So everybody
failed on him, including the Patriots because if they had any idea
how good he was, they certainly wouldn't have waited to the sixth
round, and that's a part of where I'm fascinated, and Rob, where
you and I have been talking and I've had to put this project on
hiatus, because I've been coaching my kids, but my youngest is now
a senior in high school. So I'm going to divert my attention to a
project that you and I have been discussing for a long time, like
the challenge, "Hey, is there a way to scout either college
quarterbacks or even NFL quarterbacks, possibly high school
quarterbacks?
Hugh Millen (00:56:58):
Is there a way to crunch enough data that we can have any kind of
predictive value as to their success," because if that can happen,
then you can literally write your own check because as you
mentioned, it's such a billion dollar industry and quarterback is
the hub of the wheel for any football team. We'll see how that
plays out. I did want to respond to what you said about football
and the value that it has, and I agree with everything you said,
but the way I kind of view it is this and my favorite movie of all
time is, "It's A Wonderful Life."
Rob Collie (00:57:34):
I was going to guess Pulp Fiction, but very close.
Hugh Millen (00:57:37):
Not close and there's so many great scenes and themes in that.
Remember, George Bailey, Jimmy Stewart's character. There's a point
where he's complaining to his father and I'm going to butcher this
but hopefully I get enough of the details, right? His father runs a
mom and pop hardware store at some point, George Bailey sitting
down at the dining room table, and he's saying, "Hey, I want to do
big things. I want to go build skyscrapers. I want to travel the
world, what have you. I don't want to sit back in little old Beaver
Falls, this little town and see if we can make an extra two cents
on a length of pipe," I think is one of the lines, right? Just the
routine mundane aspects of running a hardware store. So, his father
just kind of very poignantly just says, "All the people here, they
do the living and dying in this community and is it too much to ask
that they could have their own fireplace and their own roof as they
go about their mundane lives?"
Hugh Millen (00:58:31):
I put myself in that, by the way. Most of us aren't going to make a
huge impact in the world and you know, how I can prove that, tell
me the first names of your eight great grandparents?
Rob Collie (00:58:45):
Yeah, no chance. Hold on. Tom might have it.
Thomas LaRock (00:58:49):
I did a lot of genealogy. You had me stumped now, but I'll tell you
what, you gave me 10 or 15 minutes?
Hugh Millen (00:58:56):
Did you get it?
Thomas LaRock (00:58:56):
I think I could?
Hugh Millen (00:58:57):
Well, then you're in the minority?
Thomas LaRock (00:58:58):
Yeah, only because I spent a lot of time on genealogy, a lot of
that.
Hugh Millen (00:59:01):
Okay. So even if you're one of the few that could hit your eight
great grandparents, okay, now tell me your 16 great, great
grandparents. My point is just a couple generations down the line,
your descendants aren't even going to know your name. That's right,
for over 99% ... They're not even going to know your name and guess
what you might leave pictures to them, and they're going to throw
them out, because they're going to look at your picture and they're
not going to know who you are and even if you wrote, "Hey, Rob
2017," that's not going to mean anything and your descendants are
going to throw your photos out and they're not going to remember
your name, and that's true for well over 99% of us. We're just
going to live in and breathe and die and not make an impact. Was it
too much to ask that people could be excited about a Sunday
afternoon game, once a week?
Rob Collie (00:59:53):
Yeah.
Hugh Millen (00:59:54):
So that entertainment, right, wrong or indifferent, whether or not
the NFL has captivated America, the fact is, depending on how you
want to find it, for a substantial percentage of people it does and
it's something that as they're going to work, whether it's a white
collar job or a blue collar job, they're looking forward to Sunday
at 1:00 and for those people, and not everybody, there's a good
portion of the population that couldn't care less but ... and maybe
have other interests. Maybe they can't wait for the opera and if
they can't wait for the opera, God bless them and then I hope the
opera fulfills, but there's got to be something, why do we have
taxes for parks? We can do something other with that money but we
should be able to go to the park and throw a Frisbee and let our
kid slide down a slide, right?
Hugh Millen (01:00:38):
I mean, a park should be nearby and we're going to live and die and
breathe and have our great grandkids forget our first name at least
we could have a park to go to and have the Steelers to root for on
Sunday.
Rob Collie (01:00:53):
I mean, count me amongst the people, that an NFL red zone seven
hours of commercial free football coverage, the countdown music. I
actually am like, viscerally excited, watching that thing count
down like the last five or six seconds before it kicks off.
Hugh Millen (01:01:09):
Got it. Bless you. Yeah.
Rob Collie (01:01:10):
I mean, it's like Christmas morning, in a way. I'm in.
Hugh Millen (01:01:14):
Well, I saw this newscast, there's some lobster fest in Marysville
or something, whatever. Slow summer day, so they actually had a
camera out there, and this old gal, she must have been in her 90s.
She got this big smile, they asked him how the lobster is. She goes
... they're always talking about how food that's good for your
body, what's the nutrients, what's good for your body? She goes,
"Sometimes you just have to have food for your soul and lobster is
my soul food." I thought, "Wow, that's just great wisdom." So you
should have a soul activity. Hey, you look forward to that. It's
good for your soul. It makes you feel like you're living a more
honest, I won't say energized or fulfilled or whatever, but it's a
more interesting life. It's a more appealing life because you have
that. So if that's football or whatever it is for you, painting,
piano. I just think that there's endeavors like that. Most
professions aren't North Korea nuclear politics.
Rob Collie (01:02:14):
That's true. That's true.
Hugh Millen (01:02:16):
Or most diversions, I should say. The people that are providing
that for us. Maddie Damon, when he makes a movie. Thomas, right?
People in Boston call him Maddie, right? Is he changing the world?
Now, his great grandkids may remember his first name.
Rob Collie (01:02:30):
Yeah. He's the one.
Hugh Millen (01:02:31):
He's the one.
Rob Collie (01:02:32):
Generations later everyone claims to be related to him, whether
they were or not.
Hugh Millen (01:02:36):
Right.
Rob Collie (01:02:36):
So I have a personal story that underlines in business, data is
used to test or displace hunches. Decisions that have been made
because of historical tradition and even those sports, at the
highest levels have access to everything. They're still I think,
heavily, heavily, heavily influenced, especially football.
Especially American football is heavily influenced by traditional
thinking, gut instinct. So I have a cousin who, unlike me, is
incredibly athletic. He can stand with his weight evenly placed on
both feet, with his chest facing me and throw a football farther
than I can with a running start, turning sideways. Yet, whenever we
talk, he'll constantly like ... he's so humble about all of this,
right? He's like, "No, no, no. You're like, almost as good at all
of these things as I am and I am not." He has access to all of the
same information I have and he knows better.
Rob Collie (01:03:39):
He knows that we're not really the same species, but he's still
really humble about it. He's always like that. Then one day, I
asked him because he'd been a walk on receiver for some Urban
Meyer's teams at Florida, and I asked one day. I'm like, "Hey, what
is it that really separates you, Brandon, who was sixth string,
seventh string, never really saw the field, from someone like Riley
Cooper, who was out there as a starter?" I was expecting to get
some sort of like, reasonable answer about like, "Well, he's just a
hair quicker on a turn or he's a hair quicker in a decision that he
has to make," or something like that, right? I was literally
looking for that differentiator and he said, the only difference
between me and someone like Riley Cooper is opportunity. That's it.
It was such a surprise to hear this answer from him because again,
he's so humble about this all the time, right?
Rob Collie (01:04:30):
So it really stood out at me. Riley Cooper was a highly touted
recruit. The coaching staff was invested. They felt pot committed
and you see this everywhere. This is why Brady can sit the bench
behind Bledsoe. An ex NFL player recently ... I very rarely see
these sorts of things but this one happened to come by, a tweet
from Martellus Bennett. He went on this actual like series, it was
a thread of tweets. One of them was like, "You know those coaches
that you admire, half of them are idiots."
Hugh Millen (01:05:02):
Yeah. That's true.
Rob Collie (01:05:05):
You got all this analytics and all this horsepower and all this
incentive on one hand. All this pressure to break these bad habits
and yet at the center of everything, are these human beings with
these bad habits, who are in charge? I still think that some of the
biggest mysteries ... and this is just fascinating, right? Some of
the big Just mysteries about talent evaluation or about strategy or
whatever, have yet to actually happen. Some of these biggest
revolutions in it have yet to happen, because it's still ... Again,
it's that human element, right?
Hugh Millen (01:05:40):
Yeah. Well, I think that the example you cite, you could also talk
about Kurt Warner, who was the fourth quarterback in the Packers
training camp, where they had Brett Favre as we aforementioned.
Mark Brunell was the second. You have Ty Detmer was the third and
Kurt Warner. Think about Ted Bevvy and Holmgren this is a common
story. They're at a scrimmage and Kurt Warner, he had been a box
boy at a supermarket. His crew is floundering while he gets in this
camp, there's a scrimmage. At the end of the scrimmage, they're
playing another team in a scrimmage, not a preseason game but a
scrimmage and said, "All right, Kurt, your turn to get the reps,"
and he turned it down. He said, "I'm not ready." Holmgren is like,
"You don't understand this. You got to get in there, and you're
going to get four or five plays here. This is ..." and he didn't go
in and of course, he got cut and now, he's in the Hall of Fame
Rob Collie (01:06:34):
In a subsequent year when Trent Green got hurt in the preseason.
Dick Vermeil, the coach is on TV crying about their season being
ruined, at the beginning of the year before it even gets started.
He's like so disappointed for his team and all of that.
Hugh Millen (01:06:51):
You could argue that he was crying because he felt bad for Trent
Green
Rob Collie (01:06:55):
For Trent Green. Okay. All right.
Hugh Millen (01:06:57):
Because he's a crier.
Thomas LaRock (01:06:58):
That's why he was crying.
Rob Collie (01:06:58):
Come on, Hugh. Don't let the facts get in the way of the story.
Just to bring that story home for people who aren't football fans,
right? That happened, and the aforementioned Kurt Warner, they have
to turn to Kurt, the unknown Kurt. They've got to settle for the
unknown Kurt Warner and they go on, like a historic tear and win
the Super Bowl.
Hugh Millen (01:07:17):
Yeah, that very year. Yeah. I mean, it's ... if there's ever two
guys worthy of making a movie that had played in the NFL, one would
be Pat Tillman, who was an Arizona Cardinals safety and he chose to
go into the Special Forces and he was killed, walked away from $35
million to go be Army Ranger or whatever he was.
Rob Collie (01:07:36):
I know, just amazing.
Hugh Millen (01:07:37):
Right, and then the other would be Kurt Warner, because it's the
most extraordinary rags to riches story of all time, even for me, I
get nightmares about once every two months. I played 11 years and
my last conversation I had, Jim Fassel was the head coach of the
Giants and he called me and he was on his cell phone. He's driving.
He had been the quarterback coach and office coordinator at the
Denver Broncos when I had been with Denver. So he knew me and he
says, "Hey, you want to come out and be Kerry Collins' back up?" I
was like, "I'm tired of bouncing around or whatever." So, in a
technical sense, I said no, the last conversation I had, I said no.
Now, I'm haunted. I wish I would have played a few more years. Like
I said, every couple months, I have a dream that I ... I had this
opportunity to get 12 or 13 or 14, which if I had to live my career
again, I would do that because I got the rest of my life to not
play football, right?
Hugh Millen (01:08:30):
So I'm always like, I could have had 14 years but I also look back
and I piecemeal my career through and I like wait a minute, my
career might have only lasted three years because if I didn't do
this at that time, I would have been out and then there's another
point a year or two almost every single year, my career is bouncing
on the blade of the knife, so I guess the perspective is that, I
should be grateful for what I have and I am. It's human nature,
right? I always say this, Unless you were the single most
privileged person in the world which to me would be like the son of
some oil Sheik and his dad is worth 60 billion and he's got 120
foot boats and Bentley cars and helicopters and-
Rob Collie (01:09:18):
747s with pools in them.
Hugh Millen (01:09:20):
Yeah, yeah, exactly. If you somehow, somebody, you were tasked to
say, "Find me the most privileged person in the entire world out of
eight billion population," or you're the single most oppressed
person. Think of like the most oppressed ... I don't know, if he's
in a North Korea prison camp. Maybe it's some kid in India living
in a landfill. Those are the two extremes. Everybody else is in the
middle so that you can either look ahead of you and say, "Man, why
can't I be like him," or you can look below you and be grateful
that you have it better than them. It's all just, what group of
people do you want to look at? So there's only two unique people in
that discussion and we ain't one of them.
Thomas LaRock (01:10:05):
Briefly that I was going to mention ... I don't know why, but
earlier today I was thinking to myself, about how NFL drafts,
everybody gets it wrong, right? So you would think in the first
round 32 teams pick, that's their number one choice. That should be
the best player in the team in the next year or two. They should be
all pros. How many of those first rounders become all pros? They're
just so wrong about so many people. You guys, you want to talk
about predicting quarterback and I started thinking about it, and
you touched upon this right? So why do we have Tom Brady because of
one incidence, right? Mo Lewis, Marvin Lewis almost kills Drew
Bledsoe and for that we have Tom Brady and it comes back to where
... when people talk about you just get lucky, you got lucky here?
No, no luck is when preparation meets opportunity.
Thomas LaRock (01:10:53):
So whatever you guys pulled together, and I'd love to help you and
be a part of it, I've been trying to do a lot more machine
learning, predictive analytics and that's the thing that you have
to factor. It's not just the talent they have. It's their ability
to prepare, and will they be ready for the opportunity. So that's
why you have situations where a guy like Eli says, "I ain't going
to San Diego, I need to go somewhere else. San Diego is not the
right opportunity for me." Unless you have that factored in somehow
... I mean, right now on Kaggle, there's a competition they do
every year. It's called March Machine Learning Madness. So you're
supposed to build a model based upon all the previous NCAA years to
factor and come up with who rightfully should win and then apply
that model to the upcoming tournament.
Thomas LaRock (01:11:40):
I'm like, "Yeah, but what about time zones. You got a team that
travels East and has to play at noon, which is really 9 AM for them
on a Thursday. That's a huge disadvantage. I don't care what their
seat is." That's why you have a thing where Holy Cross almost be
... I forget who it was Kansas or something, right? There's so many
factors to this equation, a simple equation. Luck is when
preparation meets opportunity.
Hugh Millen (01:12:03):
Yeah, but for those of us who are interested in data, in college
football, there was a period where the computers were part of the
discernment of who was winning national championships, right? When
people say that computers as if computers are just orbiting around
Mars, just kicking out data, right? It's obviously ... it's
whatever we feed that in, a sign is important. So what you're
saying is that that needs to be factored in the algorithms, right?
That you're playing off the ... but at some point, I think those of
us who have any interest in data, we have to feel like there's
something to get back to the Tom Brady. At the very moment of draft
day, let's just use it, May 15th, 2000, all you had is what Tom
Brady had done in his life. We can agree with that, right?
Everything that we could apply about ... relevant about Tom Brady
had to have occurred up to that date.
Hugh Millen (01:13:02):
So what had he done in any way, is there anything that he had done
... For example, if you could input every conceivable data point,
"Okay, this is a guy that ... he ate oatmeal, all through childhood
instead of Froot Loops." I mean, if you could have every
conceivable variable, is there anything we were interested in data,
that we could have put into an application that could have
predicted that? I have a very fuzzy vision of the following. They
say, what would be the obvious thing that people missed about
Brady? Well, he had this penchant for making comebacks in college
when he was at Michigan, and he had a bowl game against Alabama,
where he had four touchdown pass and he brought them back from two
touchdowns back on two different occasions in that bowl game
against Alabama.
Hugh Millen (01:13:58):
They said, "Well, that's exactly like Joe Montana." Joe Montana was
a third round pick and he had this comeback against Texas, in the
Cotton Bowl. The chicken soup game if you're a big fan, because he
had the flu. So maybe you start to crystallize on that, where you
say, "Okay, those who have a pension for having those type of
comebacks, do they have a higher likelihood of being a star
quarterback?" Well, regrettably, I'm sure that what we'd find is
that there was a lot of guys who had a comeback or two, and then
you drafted them and then they didn't do anything. So what is it?
Is there something that we could have identified, that would have
said, "Hey, this thing is really spiking. Our application is
spiking. Our application is telling us that while you would think
that Tom Brady, the future six round pick that he's got like a 2%
chance of ever doing anything, this computer is actually just going
haywire and saying that's actually more like an 82% chance, that
he's going to do something.
Hugh Millen (01:15:04):
I just use pretty extreme numbers but what if it went from 2% to
32%. Now, he's not a sixth round pick, maybe and some people ...
maybe he's a third round pick. There's one thing data can tell us
after the fact but that whole realm of can you ... you guys are in
the business, you use the right term but just a predictive
component of that if we can get into that, now, all of a sudden
we're the nerds with the submarines.
Thomas LaRock (01:15:32):
I want to suck right now. My God.
Hugh Millen (01:15:35):
Yeah, you want a submarine. If you want a submarine, how about
this, you go tell Bob Kraft who's the next Tom Brady. He'll buy you
a submarine?
Thomas LaRock (01:15:42):
Actually, if I know Bob, I think he buys everything in pairs,
right? He's got the two jets. I think you'll get two submarines,
yeah.
Hugh Millen (01:15:49):
Yeah. Well, he'll give you one.
Thomas LaRock (01:15:51):
Before we sign off, I just wanted to show this the genealogy sheet.
So I knew four of the names.
Hugh Millen (01:15:56):
You knew four, four of the eight.
Thomas LaRock (01:15:59):
So here's what's funny, right, is I've got all eight because I
spoke with my grandparents while they were still alive and so I
recorded their names and then I was trying to do research. I only
have six of the 16 after that. I can't trace my history, like you
say, it's forgotten. I can't even find it. It's completely
lost.
Hugh Millen (01:16:17):
Or, you never knew it. Yeah, but the point in saying that is pretty
humbling like you guys know the blue dot, the Hubble telescope ...
that wouldn't have been the Hubble.
Thomas LaRock (01:16:27):
Pale blue dot.
Hugh Millen (01:16:27):
The pale blue dot, right? That wouldn't have been the Hobble. There
was some photo of Earth taken from halfway across the universe and
it's this little ... and you think of everything that's ever
transpired. It's all taken place on that dot.
Thomas LaRock (01:16:39):
Everybody who lived, everybody who's died.
Hugh Millen (01:16:41):
So it's humbling and I think that thinking about all the things we
worry about or whatever, to think that in two generations or no
more than three, our descendants aren't even going to know our
name.
Rob Collie (01:16:54):
Those are our descendants.
Hugh Millen (01:16:55):
Those are our descendants. Yeah, they have our blood in ...
Yeah.
Rob Collie (01:17:00):
Yeah.
Hugh Millen (01:17:00):
Yeah, our descendants.
Rob Collie (01:17:01):
Not to mention everyone else.
Hugh Millen (01:17:03):
Yeah, not to-
Rob Collie (01:17:03):
Really doesn't give a shit.
Hugh Millen (01:17:04):
Yeah, yeah. Yeah. People who owe their existence to us aren't even
going to know our names within a couple generations. It's
astounding.
Thomas LaRock (01:17:13):
What I just read, I read a quote this week that basically said,
"Just remember, no one here gets out alive."
Rob Collie (01:17:19):
Yeah. Existentialism with Hugh Millen.
Hugh Millen (01:17:23):
Okay. Yeah. Yeah. Well, we've just kind of circumnavigated and
barely touched on data.
Rob Collie (01:17:29):
This is pretty typical. I think we probably got more into data on
this conversation than on average.
Thomas LaRock (01:17:35):
Look, if we have to refund listeners' money, we will.
Rob Collie (01:17:38):
We'll give them back 10X what they paid.
Thomas LaRock (01:17:40):
Yeah, 10X
Hugh Millen (01:17:41):
At least, right?
Rob Collie (01:17:42):
Yeah, this was great. I really, really, really enjoyed this.
Hugh Millen (01:17:46):
Likewise.
Announcer (01:17:47):
Thanks for listening to the Raw Data by P3 podcast. Find out what
the experts at P3 can do for your business. Go to
powerpivotpro.com. Interested in becoming a guest on the show?
Email lukep@powerpivotpro.com. Have a data day!