Moneyball & GT Football

[Randy Carson]

Given: Our NIL fund is smaller than that of other schools; we're in the same position that the Oakland A's were in when Billy Beane implemented the now-universal Moneyball approach of finding under-valued players to build a championship team.

Question: Can Tech use statistical analysis to find 3- & 4-star players that other teams using traditional scouting methods may have overlooked and sign them for less money?

Discuss. Use both sides of the paper if necessary.

(All teams may already be doing this, but I'm just curious about the premise.)

 

[mtodd30]

Our recruiting has taken a notable uptick since NIL, in both basketball and football. I think we’re in a better position than most people think.

We might not be Ohio State, but what makes you think we have a smaller NIL pool than most of our P4 competition?

 

[gtjacket617]

We have enough NIL to win the ACC without having to play moneyball as shown in our very much improved recruiting class this year. I think this will only get better with increased wins as many Tech fans and alumni want to see a return on their investments before they bring out the big checks.

 

[gtbeak]

Given: Our NIL fund is smaller than that of other schools; we're in the same position that the Oakland A's were in when Billy Beane implemented the now-universal Moneyball approach of finding under-valued players to build a championship team.

Question: Can Tech use statistical analysis to find 3- & 4-star players that other teams using traditional scouting methods may have overlooked and sign them for less money?

Discuss. Use both sides of the paper if necessary.

(All teams may already be doing this, but I'm just curious about the premise.)

Of course, but the philosophy requires good data available to the analyst. Beane and the Oakland analysts had 7 years and 1500 PAs of data on Scott Hatteberg. I'm sure it's getting better, but I doubt that there is much good data available on the thousands of high school kids one would need to evaluate. It is likely that the concept would be more successful when deciding which transfers to target in the portal. But where this concept would be MOST successful is if one team knows what attributes to value (this is the real point of the Moneyball concept). Beane realized teams were too focused on batting average and RBI while ignoring folks like Scott Hatteberg who simply figured out a way to get on base at a high rate. Similarly I suspect that teams who focus on the currently valued attributes of players (usually skill position players) will get in bidding wars against other similarly thinking teams, while a team who figures out the hidden attributes that bring value will be the ones who are successful.

 

[slugboy]

Do we have less money than Texas? Yes.

Do we have more money than most FBS schools? Yes.

In MLB , the Dodgers are in the catbird seat. The Pirates are basically out of it. The Braves and a lot of other teams have a shot. In CFB, we are more like the Padres or Angels (mid payroll) than the A's (low,payroll).

If you want too play moneyball, what are you trying to do?

Let's look at the five factors

1. If you win the explosiveness battle (using PPP), you win 86 percent of the time.
2. If you win the efficency battle (using Success Rate), you win 83 percent of the time.
3. If you win the drive-finishing battle (using points per trip inside the 40), you win 75 percent of the time.
4. If you win the field position battle (using average starting field position), you win 72 percent of the time.
5. If you win the turnover battle (using turnover margin), you win 73 percent of the time.

How do you focus on these?

1. Explosiveness: throw deep. Get fast receivers. Air raid
2. Efficiency: flexbone. Get four yards each down. Play 1980's Washington Redskins ball
3. Drive finishing: This is a tricky one. Quoting the source article
   1. One of the most interesting aspects of writing Study Hallwas talking to coaches about this aspect of the game. Everybody agrees that it's important (obviously), and everybody had a different philosophy. When it comes to short-yardage situations, Bob Stitt of the wide-open Colorado School of Mines implements a jumbo package. Cal's air raid Sonny Dykes says, "There are a lot of creative ways to create misdirection, leverage defenders, and do the things a fullback would do."
      
      No matter how you approach it, the game of football changes when the field shrinks. It's a good problem to have -- what are we going to do once we get close to the end zone? -- but it's a question that needs an answer. The teams that make the most trips (through efficiency and explosiveness) will win more often than not, but your ability to put points on the board once in position is its own skill.
      
4. Field position: Beamer ball. Avoid sacks. Punt well. Be more aggressive on fourth down.
5. Turnovers: Tenuta ball. On offense, keep control of the ball. On defense, create havoc.

 

[GT33]

Of course, but the philosophy requires good data available to the analyst. Beane and the Oakland analysts had 7 years and 1500 PAs of data on Scott Hatteberg. I'm sure it's getting better, but I doubt that there is much good data available on the thousands of high school kids one would need to evaluate.

Beane and DePodesta’s Sabermetrics were based on measured performance metrics. What we need due to the limited performance data with HS players is a composite physical and mental aptitude measurements. I’m an expert, but physically it’s the small ankle kids, agility freaks and those with high motors. There’s a lot more data with baseball with all the games these kids play.

 

[MacDaddy2]

Do we have less money than Texas? Yes.

Do we have more money than most FBS schools? Yes.

In MLB , the Dodgers are in the catbird seat. The Pirates are basically out of it. The Braves and a lot of other teams have a shot. In CFB, we are more like the Padres or Angels (mid payroll) than the A's (low,payroll).

If you want too play moneyball, what are you trying to do?

Let's look at the five factors

1. If you win the explosiveness battle (using PPP), you win 86 percent of the time.
2. If you win the efficency battle (using Success Rate), you win 83 percent of the time.
3. If you win the drive-finishing battle (using points per trip inside the 40), you win 75 percent of the time.
4. If you win the field position battle (using average starting field position), you win 72 percent of the time.
5. If you win the turnover battle (using turnover margin), you win 73 percent of the time.

How do you focus on these?

1. Explosiveness: throw deep. Get fast receivers. Air raid
2. Efficiency: flexbone. Get four yards each down. Play 1980's Washington Redskins ball
3. Drive finishing: This is a tricky one. Quoting the source article
   1. One of the most interesting aspects of writing Study Hallwas talking to coaches about this aspect of the game. Everybody agrees that it's important (obviously), and everybody had a different philosophy. When it comes to short-yardage situations, Bob Stitt of the wide-open Colorado School of Mines implements a jumbo package. Cal's air raid Sonny Dykes says, "There are a lot of creative ways to create misdirection, leverage defenders, and do the things a fullback would do."
      
      No matter how you approach it, the game of football changes when the field shrinks. It's a good problem to have -- what are we going to do once we get close to the end zone? -- but it's a question that needs an answer. The teams that make the most trips (through efficiency and explosiveness) will win more often than not, but your ability to put points on the board once in position is its own skill.
4. Field position: Beamer ball. Avoid sacks. Punt well. Be more aggressive on fourth down.
5. Turnovers: Tenuta ball. On offense, keep control of the ball. On defense, create havoc.

Also worth noting that the team with the better recruiting wins 75% of the games. The team with better recruiting and coaching wins 85% of the time.

 

[GT33]

Also worth noting that the team with the better recruiting wins 75% of the games. The team with better recruiting and coaching wins 85% of the time.

We need to dominate the 15% that supposedly wins with neither

 

[Vespidae]

Yawn.

This was the whole strategy of TStan. Find 3 stars and develop them into 5 stars. It’s still a reasonable approach. The constraint is Tech’s academics and limited curriculum.

Which means we need to be world class in recruiting and development.

 

[MountainBuzzMan]

Given: Our NIL fund is smaller than that of other schools; we're in the same position that the Oakland A's were in when Billy Beane implemented the now-universal Moneyball approach of finding under-valued players to build a championship team.

Question: Can Tech use statistical analysis to find 3- & 4-star players that other teams using traditional scouting methods may have overlooked and sign them for less money?

Discuss. Use both sides of the paper if necessary.

(All teams may already be doing this, but I'm just curious about the premise.)

What is our NIL budget compared to other schools? We have a smaller fan base but a much wealthier fan base. We could very well be in the top 10-15. But our fanbase is also not foolish with money. This is why I really like having Key. He and his hires seem like a really good investment. Which bodes well for punching way above our weight from an NIL perspective.

 

[cpf2001]

The focus on line play could be a bit of this. Less flashy than top-end skill player transfers but maybe more long term bang for the buck.

The staff also seems good at identifying QB prospects.

 

[RonJohn]

I don't think there is enough data to make the statistical analysis valuable.

First look at college football vs MLB. MLB teams have many more games per year. MLB teams had played the game much the same, with the same assumptions for probably at least 100 years before Oakland started looking at things differently. MLB keeps detailed statistics about the games and players, and have for more than 100 years. That means there was a very substantial amount of data available to complete statistical analysis on. College football keeps some statistics, but they are not as detailed as MLB. How many times per year in MLB is there an opportunity for a double play with no outs while a runner could score if you complete the double play? How many times per year in college football is there a 4th and 4 on your own 42 yard line? How many times have the college football teams gone for it on 4th and 4? I don't know, but I think there would be a large difference in the number of times that MLB teams hold the runner vs attempt the double play. More data, and more varied data results in better statistical analysis with which to make decisions.

Secondly, how much data is available about high school recruits? I would say almost none as a comparison to MLB statistics. You know the height and weight. Maybe some measurables, but those measurables are not strictly controlled. The high school athletes do not go to a single event in a central location to be measured. Is a 0.1 second difference in 40 time, or three cone drill because one athlete is faster, or because one guy is slower with the clock? Does a high school RB run for 1,000 yards because he is extremely good, or because the region his high school in is not very talented? Is it possible for an extremely talented high school RB to only gain 600 yards because his teams offense only needs him to run occasionally? But he is an extremely good pass blocker, and gains a first down every single 3rd down that he is asked to run?

For statistical analysis to help, you need enough statistics to get an accurate result. ESPN has started doing "analytics" during games. I am not very impressed with the win probability number they use. It seems to me like they could have a guy watching the game and updating that based on his opinion, and it would very closely match their number. They started last year showing "ESPN Analytics" for whether a team should go for it on 4th down. I saw some where it said a team should not go for it on 4th and 3 or less from the other teams 40 yard line. I saw some where it said a team should go for it on 4th and less than 7 on their own 35 yard line. (I don't remember the actual numbers, but I did see some in that range.) I see no way that could be accurate. I'm sure some will say that MLB managers didn't believe the moneyball analytics until it was used. I would counter that MLB managers could have looked at the data and the analysis methods used to decide whether it made sense or not. We do not know what data ESPN is using for the "ESPN Analytics". We do not know what statistical methods ESPN is using for those. I am not going to put faith into their numbers just because the call them analytics.

 

[takethepoints]

On the stats question: yes, and think the situation is actually worse than Ron is pointing out. The kinds of stats that sabremetrics uses are more detailed in situations then he is saying. It's all the way down to how a player reacts in certain situations and certain positions on the field when the ball is landing in particular areas of the field and what the results are. Now, you could do this for high school players, but the results would probably not be helpful until their junior year at the earliest. And that really cuts down on the sample size of plays that go a player's way. This is one reason that Beane and co. always preferred to draft undervalued minor league and college players; more data. And this leaves aside the problem that collecting valid data from high school film would be extremely difficult; it's lot easier to see how a player gets hits then if his blocking on the interior line is effective. So we end up with, at best, ranks for players based on selective data concerning each player; i.e. the "star" system. It's better then nothing for a reason. Getting the right kind of data would be extremely expensive and with no guarantee that the end result would justify the cost.

On Moneyball itself: As I pointed several times when Paul was coach, Tech was playing Moneyball. What the system he ran did for Paul's recruiting was make the roles of specific players on offense clear as crystal and, as a consequence, led to easier decisions about what player would fit where. And it worked. All the system needed to make real hay was a QB who could both run and pass efffectively. For the other positions it was a matter of finding player others didn't want or didn't value correctly. Soooo … when most teams wanted Mills for LB, Tech offered him the premier back position and got him. Same for Peebles; "too small" most teams said. Same for Shaq; "too short" most teams said. And so on. System offenses make recruiting a lot easier. And Tech was always in the top 10 - usually top 5 - in rushing. This didn't work for D recruiting, of course, but when the O was running well, the D sat on the sidelines drinking Gatorade and all was well.

Now Tech is running an O that is even more QB dependent with the same kind of D everybody else is working with, but more money and help with recruiting - the staff is pretty good at this and is numerous - appears to be paying dividends. But Moneyball it ain't.

 

[SOWEGA Jacket]

Right premise and we are doing it by spending on OLinemen. In my 20’ and 30’s I was all about the flash of WR’s and RB’s. I now understand that it’s a numbers game not a flash game. GT is spending money on positions that are hard to find. There are thousands of RB’s and WR’s that are all within a small window of talent. That’s not the case for OLinemen. And without OLinemen you could have a Heisman contender at QB and a Heisman winner at WR and not win anything meaningful (see what I did there).

Sure, it would be nice to have 5 star Linemen and 5 star WR’s and RB’s like Bama, UGA, and Ohio State had over the years but those are the exceptions. Most teams in the middle to above average tier will win based on their OLine, QB, and coach. We saw that under Johnson and Gailey. Gailey had star talent and was saverage. Johnson was laughed at in recruiting and won many games.

Key is building this team the right way and is playing Moneyball by letting a guy like Singleton go so he can use that money elsewhere. The next step is for Key to become a better coach himself or he’ll be another Gailey. Key has been in the game a long time so, like Johnson, it’s time he shows us what he’s learned regardless of losing Santucci or when we lose Faulkner or King. I want him to succeed so badly and I’m excited how we are handling this new era. The portal and NIL are the best recruiting tools GT has ever been handed. We know longer have to play defense about academics or “calculus” since academics no longer matter. Now, we just gotta spend wisely and get some studs to want to live in Atlanta - which he is doing.

Now, don’t lay an egg on national TV in Boulder.