Tomlin & FO have staff that crunch numbers and do analytics.

[jeemie]

I think in the NFL analytics could be useful if limited to actual events where you can be reasonably certain the numbers are actually measuring something the player did and not something that happened because it looked like he did something wrong, but actually another person blew their assignment.

Also important to keep in mind the situations in which those measurements took place, so the numbers have context.

I really like the way Pat Kirwan integrated metrics into actually scouting players and watching film on players (detailed in Take Your Eye Off the Ball).

He would use numbers like Explosion Quotient (a combine-generated number) and Production Quotient (a number generated by plays a player actually made in games) together with the situations in which those players were used to come to a very well-rounded judgement as to whether a player had the potential to be good. I liked this because his metrics usually were based off measurements that could be isolated to the actual player it was supposed to be measuring, and he also looked at the real-world context into which the numbers fit.

[Kodiak]

However- scientists have started to show...using data...that a lot of what you think of as a conscious choice was actually a "decision" made in your subconscious mind LOONG before you become consciously aware of this...and if you study patterns of human behavior (and analytics is pretty much the study of patterns in data) you can be clued in to what some of these "sub-conscious" choices might be...like what Netflix does.

Being more self-aware doesn't help with these types- in fact, so much of what we think of as "self-awareness" may very well not be.

That's a better way of putting it. People don't know what they don't know, and often they don't know what they think they know, either.

If you're company isn't using advanced stats/data, it's not reaching even 70% of its potential (assuming it even stays in business). I can only imagine what people said about SPC in its early days.....any student entering college today regardless of what they study, they should include several classes in programming and statistics.

Advanced stats hold people accountable, that's why so many rebel against and deride them. Quite simply it's the bane of underperformers.

[Legacy User]

This thread is instructive.

The fact remains though that Gut>Stats, and Gut>Math

Seriously. If you follow the stats you would never have Sushi try 50+ yard field goals with two minutes left in a tie game. Or call timeouts after the two minute warning instead of before the two minute warning in a game we're losing. Only a great man's gut can come up with that kind of stuff.

Case closed.

Freakin' nerds

[jeemie]

This thread is instructive.

The fact remains though that Gut>Stats, and Gut>Math

Seriously. If you follow the stats you would never have Sushi try 50+ yard field goals with two minutes left in a tie game. Or call timeouts after the two minute warning instead of before the two minute warning in a game we're losing. Only a great man's gut can come up with that kind of stuff.

Case closed.

Freakin' nerds

I took a lot of grief from Swiss over my insistence that Tomlin never should have trotted Suisham out there for that 52 yarder in the 2012 Tennessee game...and my argument was a stats-based argument coupled with a situational game management argument. At that point in his career, he had attempted only 11 of 167 kicks from that range, and had hit only on 4 of them- one of those four he had just hit earlier in the same game! So Tomlin was counting on him to hit TWO 50+ FGs in the same game! But Swiss was saying all those numbers should be ignored because maybe Tomlin could see Sweezy was "booming them all night" and he "felt good".

But resting your hopes on a 50+ yard FG with plenty of time left, a shaky defense, a kicker on the other team who IS an historically good long-range kicker (indeed he was an historically consistently good FG kicker, period) where a missed FG gives the opponent excellent field position was just bad risk management.

[zeke5123]

However- scientists have started to show...using data...that a lot of what you think of as a conscious choice was actually a "decision" made in your subconscious mind LOONG before you become consciously aware of this...and if you study patterns of human behavior (and analytics is pretty much the study of patterns in data) you can be clued in to what some of these "sub-conscious" choices might be...like what Netflix does.

Being more self-aware doesn't help with these types- in fact, so much of what we think of as "self-awareness" may very well not be.

That's a better way of putting it. People don't know what they don't know, and often they don't know what they think they know, either.

If you're company isn't using advanced stats/data, it's not reaching even 70% of its potential (assuming it even stays in business). I can only imagine what people said about SPC in its early days.....any student entering college today regardless of what they study, they should include several classes in programming and statistics.

Advanced stats hold people accountable, that's why so many rebel against and deride them. Quite simply it's the bane of underperformers.

I am not disagreeing with this. I would just caution against the belief that advanced stats can be used to predict the future, because of the inductive reasoning problem. You need to always be a bit conservative and understand the downside risks. I think the more important skill is learning how to falsify. Too often, we are obsessed with proving what is right. It can be more important often to prove what is wrong and in that way cut excess.

[Legacy User]

However- scientists have started to show...using data...that a lot of what you think of as a conscious choice was actually a "decision" made in your subconscious mind LOONG before you become consciously aware of this...and if you study patterns of human behavior (and analytics is pretty much the study of patterns in data) you can be clued in to what some of these "sub-conscious" choices might be...like what Netflix does.

Being more self-aware doesn't help with these types- in fact, so much of what we think of as "self-awareness" may very well not be.

That's a better way of putting it. People don't know what they don't know, and often they don't know what they think they know, either.

If you're company isn't using advanced stats/data, it's not reaching even 70% of its potential (assuming it even stays in business). I can only imagine what people said about SPC in its early days.....any student entering college today regardless of what they study, they should include several classes in programming and statistics.

Advanced stats hold people accountable, that's why so many rebel against and deride them. Quite simply it's the bane of underperformers.

Statistics are exactly as valid as the data is good. In the NFL, sales or anything you can wander into a shitty space pretty quickly with bad data... Or even find yourself doing nothing because you have people who are supposed to be doing something reliving their past gathering data.

Statistics a pretty clear that teams should never punt. If Tomlin embraced that this board would explode.

[jeemie]

Statistics a pretty clear that teams should never punt. If Tomlin embraced that this board would explode.

I wouldn't...and I don't think Gambler's Ruin applies to this strategy either (that's a popular objection to the never punt argument) but haven't quite formulated my rebuttal yet.

[Legacy User]

You're rebuttal really doesn't matter. People aren't rational- that's part of the "human element".

Many times it matters not whether you are right or wrong, but only if you succeed the first time you try something novel.

This how the world works.

[zeke5123]

Statistics a pretty clear that teams should never punt. If Tomlin embraced that this board would explode.

I wouldn't...and I don't think Gambler's Ruin applies to this strategy either (that's a popular objection to the never punt argument) but haven't quite formulated my rebuttal yet.

I am unsure what you mean by Gambler's Ruin in this context.

The real issue most of these guys don't seem to understand is the diminishing marginal return of points, which means optimizing EV is probably a bad idea. Now, coaches are probably too conservative on 4th downs as is, but going for it all the time probably isn't optimal either once you account for diminishing marginal returns.

[lifelongsteel]

I'm quite impressed with the board knowledge on this subject.

At a broader level, I see three types of responses to stats/analytics across front offices regardless of sport:
- Enthusiasm: we can get an edge
- Open mindedness: let's see if this helps and we'll then get on board
- Skepticisim: it will never work

It is clear in other sports, most notably baseball but basketball more and more, that the teams with the enthusiasm for this are gaining a sustainable edge over the other two groups. The reason is that the general purpose stats that often get published, don't get published or widely understood until they are already outdated. You need to be out front on this, generating and testing your own hypotheses, building your base of knowledge etc. For example, people thought Billy Beane was crazy giving away secrets in moneyball. The reality was that he and his staff had long moved on to other fields of study. Other places to find an advantage.

I try not to read too much into what Tomlin says in public, but combining his comments with what i see with my own eyes, I would put the steelers in that second group. I believe that in 3-5 years it will come to light that the Seahawks, Patriots, Eagles and a few others have a built in advantage over other teams due to the investment they made in this field. We will not be one of those teams.

[Legacy User]

However- scientists have started to show...using data...that a lot of what you think of as a conscious choice was actually a "decision" made in your subconscious mind LOONG before you become consciously aware of this...and if you study patterns of human behavior (and analytics is pretty much the study of patterns in data) you can be clued in to what some of these "sub-conscious" choices might be...like what Netflix does.

Scientists have nothing to say about consciousness or subconsciousness because they study neither. They study a physical organ, the brain. What neuroscience is starting to show is that certain physical markers in brain temporally precede decisions (I find this fact to be most irritating!); that suggests that the brain is the cause of decisions not that decisions cause physical reactions in the brain. But whether those markers precede thought or not, who knows. Still early, early days.

I would cautiously point out that THIS

Advanced stats hold people accountable, that's why so many rebel against and deride them. Quite simply it's the bane of underperformers.

DOES NOT WITH NECESSITY IMPLY THIS

You cannot regularly succeed without reliance on advanced stats.

I'm happy the Steelers are investing in analytics. As Kodiak has very nicely pointed out, I hope they are investing in the right people to do the job and that those in a position of influence are wise enough to listen.

I believe that in 3-5 years it will come to light that the Seahawks, Patriots, Eagles and a few others have a built in advantage over other teams due to the investment they made in this field. We will not be one of those teams.

Life, I would not be surprised were this to be the case.

[zeke5123]

Also, from what I have read many of those experiments with the brain and decision-making dealt with rudimentary things, such as eating a muffin, etc. I have doubts that the same is true for more complex issues.

[jeemie]

Lit...in those studies, scientists could predict a person's decision ~10 seconds before the decision maker was consciously aware of the choice he made...with very high accuracy.

So it was more than just "the brain lit up" before the decision was made.

It was that a CERTAIN PART of the brain lit up depending on what the ultimate choice was and scientists could guess the decision before the decision maker was aware of it.

[jeemie]

Also, from what I have read many of those experiments with the brain and decision-making dealt with rudimentary things, such as eating a muffin, etc. I have doubts that the same is true for more complex issues.

I bet it's not that much different.

[Legacy User]

Lit...in those studies, scientists could predict a person's decision ~10 seconds before the decision maker was consciously aware of the choice he made...with very high accuracy.

So it was more than just "the brain lit up" before the decision was made.

It was that a CERTAIN PART of the brain lit up depending on what the ultimate choice was and scientists could guess the decision before the decision maker was aware of it.

Oh, I know. I was just offering a corrective. Neuroscientists study brains, not consciousness. But whether a physical state in the brain precedes thought and whether thought is the same thing as a physical brain state or merely requires the brain state, these questions are still very much open. For instance, if you identify thought as being nothing other than a physical brain state, then all the study shows is that a brain state precedes a brain state. But it is very irritating to discover that certain forms of thought may be the result of a physical state and not the other way around!

I think the NFL should hook some electrodes up to Tomlin's brain during games so we can start groaning 10 seconds before the decision is made.

[Legacy User]

I'm quite impressed with the board knowledge on this subject.

At a broader level, I see three types of responses to stats/analytics across front offices regardless of sport:
- Enthusiasm: we can get an edge
- Open mindedness: let's see if this helps and we'll then get on board
- Skepticisim: it will never work

It is clear in other sports, most notably baseball but basketball more and more, that the teams with the enthusiasm for this are gaining a sustainable edge over the other two groups. The reason is that the general purpose stats that often get published, don't get published or widely understood until they are already outdated. You need to be out front on this, generating and testing your own hypotheses, building your base of knowledge etc. For example, people thought Billy Beane was crazy giving away secrets in moneyball. The reality was that he and his staff had long moved on to other fields of study. Other places to find an advantage.

I try not to read too much into what Tomlin says in public, but combining his comments with what i see with my own eyes, I would put the steelers in that second group. I believe that in 3-5 years it will come to light that the Seahawks, Patriots, Eagles and a few others have a built in advantage over other teams due to the investment they made in this field. We will not be one of those teams.

With 162 games and playing the statistical odds makes a lot of sense. A 16 game season just doesn't have the sample size to play statistics the same way. Just too many variables in weather, the field conditions, injury and fatigue on both sides of the ball that baseball doesn't encounter.

[lifelongsteel]

I'm quite impressed with the board knowledge on this subject.

At a broader level, I see three types of responses to stats/analytics across front offices regardless of sport:
- Enthusiasm: we can get an edge
- Open mindedness: let's see if this helps and we'll then get on board
- Skepticisim: it will never work

It is clear in other sports, most notably baseball but basketball more and more, that the teams with the enthusiasm for this are gaining a sustainable edge over the other two groups. The reason is that the general purpose stats that often get published, don't get published or widely understood until they are already outdated. You need to be out front on this, generating and testing your own hypotheses, building your base of knowledge etc. For example, people thought Billy Beane was crazy giving away secrets in moneyball. The reality was that he and his staff had long moved on to other fields of study. Other places to find an advantage.

I try not to read too much into what Tomlin says in public, but combining his comments with what i see with my own eyes, I would put the steelers in that second group. I believe that in 3-5 years it will come to light that the Seahawks, Patriots, Eagles and a few others have a built in advantage over other teams due to the investment they made in this field. We will not be one of those teams.

With 162 games and playing the statistical odds makes a lot of sense. A 16 game season just doesn't have the sample size to play statistics the same way. Just too many variables in weather, the field conditions, injury and fatigue on both sides of the ball that baseball doesn't encounter.

If you knew that something statistically had a 60% chance of working you would choose it regardless of the number of opportunties you would get to make that choice.

It's like playing blackjack . . . even if you're playing one hand it's probably not a good idea to hit with 14 staring at a dealers 6.

[jeemie]

I think the NFL should hook some electrodes up to Tomlin's brain during games so we can start groaning 10 seconds before the decision is made.

This is awesome.

[Legacy User]

I think the NFL should hook some electrodes up to Tomlin's brain during games so we can start groaning 10 seconds before the decision is made.

Right idea. Wrong body part.

Gurgle. Burp. Challenge Flag.

That's how it works.

[Gonzo]

The advance stats are to find value - value in personnel choices, trends and philosophy/strategy

You could argue the initial success the steelers had with the 3-; wasn't scheme so much as personnel choices by gaining value moving DTs to DEs DEs to LB etc ...

Innovation can be accomplished by statistical analysis that can help shape game philosophy and and player selection.

Stats/Trends can help spot future advantage more so IMO than simply a making a play decision.

I don't think Tomlin ignores simple stats like 4th down numbers, kicking percentage etc .... I think he has an unflappable belief in ass-slapping and a misguided idea of attrition football

[lifelongsteel]

The advance stats are to find value - value in personnel choices, trends and philosophy/strategy

You could argue the initial success the steelers had with the 3-; wasn't scheme so much as personnel choices by gaining value moving DTs to DEs DEs to LB etc ...

Innovation can be accomplished by statistical analysis that can help shape game philosophy and and player selection.

Stats/Trends can help spot future advantage more so IMO than simply a making a play decision.

I don't think Tomlin ignores simple stats like 4th down numbers, kicking percentage etc .... I think he has an unflappable belief in ass-slapping and a misguided idea of attrition football

That's a great post with some Lit quality verbiage. "unflappable belief in ass-slapping" is quite good.

[Gonzo]

Tomlin says - "crunch this, if you will" and then high fives everyone

[Legacy User]

Ha ha. Quality posts, Gonzo.

[jeemie]

What is the three semicolon defense, Gonzo?

[Gonzo]

What is the three semicolon defense, Gonzo?

I am a lazy typist
3-4
Or not,
It's Tomlins Version of the 3-;) if you will

[jeemie]

The Three Wink Defense, obviously.