"What If" on the coaching front?
Now that the NBA finals are over and sports enters the dark period know as baseball season, I am trying to get back in the swing of Michigan football. There has been a lot of grumbling recently about our recruiting of non-blue-chip recruits. The 3-9 season with losses in the top-flight rivalry games, loss of actual or expected impact players via transfers, and (to a much lesser extent) the aspersions on Rodriguez's character (see, e.g., snake oil, lack of family values) have also caused some heartache for Michigan fans.
So here's my question, if you wound the clock back and could choose between Les Miles and Rodriguez (I'm not saying this was the situation we actually found ourselves in), who would you choose?
Personally, I am drinking the Rodriguez Kool-Ade. I am sold on the spread as an offensive system and think that it will pay dividends once we get the right personnel installed. I was a fan from the beginning of Barwis and the long-overdue revolution in the weight room. Rodriguez’s track record is obviously stellar even though he isn’t a Michigan Man in the traditional sense. I am confident that greatness will ensue in the next year or three.
I was also, however, intrigued by Les's 4th down gumption and success at LSU. I have no clue if the transition with Coach Miles would have been smoother. Would we have performed better last year and maybe lost fewer players to transfers if we weren’t installing the spread on the fly? Would we be recruiting four and five star players in droves right now? I don’t know. If it was your pick, who would you have at the helm of the leaders and best?
hot dog! we have a wiiiener!
I mostly appreciate any and all sports. The only sports that I'm "in to" are baseball, football, basketball, hockey, and wrestling. I find soccer to be incredibly boring. That doesn't mean I don't appreciate the intricicies and hard work the players put in.
Miles is an odd person...strange...would he have really been a good fit, in spite of being a Michigan Man?
I respect him as a coach, but I was always uneasy about the fact that Lloyd Carr was adamantly opposed to his hiring. If he indeed spread rumors about Carr's health on the recruiting trail, that was pretty bush league (not to mention stupid, given that Carr was going to be influential in the hiring process). It may not be a fair assessment, but Miles gives off a kind of shady, "SEC" vibe.
Here's the problem. Les Miles was not the only head coach out there spreading rumors about Lloyd Carr's health. Allegedly, Kirk Ferentz was doing the same damn thing, yet, Carr would've wholeheartedly endorsed Ferentz as his successor.
but KF never banged LCs wife.
allegedly!
Well, you're kind of making my point for me. The animosity that Carr and Miles feel towards eachother probably isn't football related.
Where did you hear that Ferentz used Carr's health against him? I've never heard that, and it really doesn't seem to fit his character at all.
That's not the only knock against Les, anyway. Miles was also reputedly reprimanded by Bo for committing recruiting violations when he was an assistant here. He was subsequently accused of similar funny business during the recruitment of Jai Eugene. As for the "Loose Morals" stuff that allegedly went on (though I've heard it was Moeller's wife, not Carr's), well, that didn't help his cause, either. I suppose it's possible that Miles was unjustly accused of these things. But frankly, it doesn't seem like it would be out of character for him have done so.
I don't know if things would be better if Coach Miles were the man here at Michigan. The 4th down play was/is thought of as insanely risky at best, and that is after the fact that it succeeded. Had it failed, he would have burned in LSU football purgatory. He's a gambler and I don't think that philosophy would fare well against Coach Tressel's style, a guy who focuses on fundamentals and has some mighty good talent to work with as well. It might take Coach Miles longer to beat Coach Tressel than Coach Rodriguez will for all we know. Right now, I feel good about how things worked out even with the 3-9 season.
I never thought Miles was as good a coach as others apparently did. He did an amazing job with a large number of legacy guys from Saban, and his "gumption" on 4th down could have easily blown up in his face. I am all for a coach going for it on 4th down with about a yard or 2 to go, but he went for it often enough, in situations that are not really that advisable, that I think he just got lucky on a few of them. Against good teams, most plays are 60/40 or 50/50 in terms of success, and that is true on 4th down as well. Miles may have been flipping heads quite a bit, but at some point the law of averages will start to work against him, and his gambling may start to look more reckless.
That said, I think he would have been a decent coach here at UM, but I think that, in the long run, RR will work out to be a better choice.
It's only gambling if your odds are 50% or less. Additionally, there is no such thing as a law of averages.
From Wiki "As invoked in everyday life, the "law" usually reflects bad statistics or wishful thinking rather than any mathematical principle. While there is a real theorem that a random variable will reflect its underlying probability over a very large sample, the law of averages typically assumes that unnatural short-term "balance" must occur."
I understand that the "law of averages" is not a real scientific law, but your own quote affirms the underlying principle - that a "random variable will reflect its underlying probability over a very large sample size." The issue here is that 4th down conversions are, by their very nature, relatively few in number for a given season. But if you look at the 4th down statistics from last year nationally, you see that the rate is about 50% (or what you referred to as "gambling"). So teams are taking a gamble when they try these plays, and it is basically a coin flip if they work.
http://www.ncaa.com/sports/m-footbl/stats/ncaa-m-footbl-fbs-team-4th-do…
In fact, LSU was below this conversion percentage last year (38%), but I suspect that they've had at least one season with an above-average conversion rate (I could not find statistics outside of 2008). Now, I know that there are different levels of 4th down conversions (down 3, minute left vs. 4th and 1 in the second quarter), but just from experience it seems like teams fail about as often as they succeed.
But you say that it is not "gambling" unless you have a =50% chance of conversion; at the same time, though, you fail to recognize that sample size plays an immense role in it being a "bet" or not. That's why Vegas casinos are cash cows - a gambler might have a run of luck, but unless you "cheat the system" (i.e. card count like MIT kids, and even then they were ultimately caught and, in a sense, "lost"), you'll lose more than you win. The games are simply set up to favor the house.
But I know that statistics can be deceiving - Sure, teams like UCLA and Florida Atlantic probably had horrible conversion rates because they are not great teams, but shouldn't good teams have a higher-than-average conversion rate? That I'm not sure of, but I suspect that, normalized for opposing talent, everyone's 4th down percentage is probably close to 50%, or a bit under. Going for it on 4th and 2 against Toledo up 20 will be quite a bit easier than against OSU down 3. I'm not saying that Miles is wrong for going for it on 4th down in certain situations, but history shows that 4th down conversions are a dicey proposition, and a coach who succeeds at a high rate for a small period of time will likely come back to the pack a bit as the sample size goes up.
"But you say that it is not "gambling" unless you have a =50% chance of conversion; at the same time, though, you fail to recognize that sample size plays an immense role in it being a "bet" or not. That's why Vegas casinos are cash cows - a gambler might have a run of luck, but unless you "cheat the system" (i.e. card count like MIT kids, and even then they were ultimately caught and, in a sense, "lost"), you'll lose more than you win. The games are simply set up to favor the house. "
If I'm > 50% to win, unless I can't afford to lose (this would hold REGARDLESS of the odds, it requires a minimum level of utility to be present in any outcome. Basically, if you can't afford to lose, you can't bet even if you're 99.9% to win.), I would make the same bet whether I get 1 trial or 100. Also, memoryless property and all, if I win the first 3 of 4 bets, the fact that I won the first 3 doesn't change the odds at all.
TL;DR - The odds favor the house in the long AND the short run.
That's fine if you are more than 50% likely to win, but as I showed conversion percentage for teams in 2008 was less than 50% (49.09% to be more exact). And with gambling, losing money tends to be less dramatic on a per-hand basis - you may lose this hand, but it doesn't really affect the next hand except your odds might rise or fall slightly with certain games (most commonly in blackjack that does not feature per-hand reshuffling). Provided you have enough funds to suffer some losses, you can go right back to betting with, as you say, memoryless property. In football, failing to convert on 4th down will give the other team the ball and, depending on the circumstances, better field position and/or the ability to run out the clock. That result has a memory, and will affect what you are able to do in both the near and far futures. In football, even if you have better than 50% chances of winning, you cannot always afford to lose. That's where I have the biggest disagreement with your argument - you act as if every football play in worth the same; 1st down in the first quarter has as much "effect" on the game as 4th down with 2 minutes left. I know that you wouldn't agree with that statement, but that's what you are proposing - grind it out, take your losses, the numbers will work out. But converting on 4th down, as we've seen both in college and the NFL, is basically a coin flip.
Clearly Miles shares your view on 4th down - that if you honestly "feel" like you have better than 50% odds on converting, go for it. But when you are not actually converting at that rate, that can be a major factor in going 8-5 and losing some tight games.
And as for your argument that the odds always favor the house in the short and long term, that is true. I said that. So no complaints there. But as I showed in 2008 (and I would love to see more stats for other years), going for it on 4th down also favors the defense in both the short and long term too.
"If I'm > 50% to win, unless I can't afford to lose (this would hold REGARDLESS of the odds, it requires a minimum level of utility to be present in any outcome. Basically, if you can't afford to lose, you can't bet even if you're 99.9% to win.), I would make the same bet whether I get 1 trial or 100."
Every play isn't worth the same, that's what I was trying to convey here.
Also, we're mostly on the same page, but I forgot to include consideration for the expected value of the outcomes. You're going to give the ball over by punting if you don't go for it anyway. I guess a better argument from me would have been to note that once you're at a certain field position, punting is worth less than going for it.
Now that I agree with. I actually think that going for it once you are past the 35 yard line or so, to an extent, is the better option that kicking a FG, at least in college.
There are two areas where hiring Miles would have been more beneficial. The first is in Michigan high school recruiting. A Miles hire would probably have stopped Dantonio from at least winning the perception war--that Dantonio is ruling the state and Rodriguez isn't interested in Michigan talent. The second is in the area of so called "negative recruiting"--which is a nice catch phrase, reallying cry, and simple justification for guys who are Scout cult members. But I do think there is some truth to that--competing coaches for recruits that UM is after are probably telling these recruits that people at and in Michigan don't even like Rodriguez and his days are numbered (easier to pull of after 3-9).
And if UM would have snagged Miles AFTER winning an NC at LSU--that would have added to his cache, and his ability to recruit. The transition would have been much smoother. I believe Mallet and Boren stay, and possibly Arrington as well as a couple of the other offensive linemen who quit. I think Miles would have brought serious swaggerrooni with him, and I love his penchant for getting hellified assistants as well. At Michigan, Miles would be as close to a transcendental figure as you can get--and that's what I think UM needs.
Rodriguez needs something which he may not have: TIME. This is the hire--you have to go with it, and you MUST support it eleventy billion percent. There is no other choice. And I do think that with time he can become a transcendental figure by overcoming all the BULLSHIT he has had to deal with, whether the BULLSHIT was his choice or dickheads in the Detroit media, or the elitist "Michigan Man" snobs in Ann Arbor. If he can dig his heels in, be a bit of a rapscallion, a bit of a "I'll show you, you fuckers, you" and win--and win the fans, media, and snobby fags over, then he just might be the guy who can enter into the Realm of BO.
I am not so sure it will be done--but if he is given time, and a lot of these shlub recruits are "chip on their shoulder" types who work their asses off and show that Rodriguez DO have the eye for talent we are all praying to Mighty Zeus for, this thing will turn out wonderfully.
Last point before I die (although, to my knowledge, I'm not dying--but you never know): Rodriguez is a prototypical SEC type coach. He was born to be in that league. We have him here, and I won't be surprised if he turns into a total ass kicker and take no nameser. I write this just after he has plucked Courtney Avery from the Drunk Punk of Stanford--and I for one extend both middle fingers in saluting Jim Harbaugh and laughing at him.
is that if we had gotten Les Miles as a coach, he wouldn't have a NC ring. If you recall, he was in discussions for the Michigan job BEFORE the bowl season. Minor detail.
You summed up my thoughts pretty well, but the RR as Bo scenario might be a stretch. I hope to hell not, time will tell.
I think you have to look at 4th down calls as an expected value situation. It's not just a matter of "how likely are we to convert this?" or "can we afford to lose this?" You can afford to fail to convert any fourth down in the first quarter, but you might not want to risk handing your opponent easy points by going for it inside their field goal range even if you think you have a great shot at converting. You're probably more interested in trying to convert a fourth down from just outside your own field goal range because a conversion is more likely to result in points.
I feel like the more interesting aspect of play-calling is the paper-rock-scissors aspect of things. You can chart and diagram all you want. I am sure it's useful to be cognizant of what plays work best at what down and distance. But with all the moving parts and variables in a football game, so much of play-calling is a gut feeling. Something like the hook-and-ladder that sealed the Purdue game last year was just a great call. I don't know if you would ever cue that up based on stats alone, but it was a devastating, fantastic call.
What makes certain coaches good and creative with that aspect of the game? Is it luck, instinct, or just amazing scouting and film study to identify weaknesses in a defense?
What makes coaches good and creative with that aspect of the game? I am impressed with your use of a thesaurus and Big Terms like "expected value situation." Nice.
If "expected value situation" is a synonym for "wants to win"--then I agree.
Otherwise, what makes a coach "good and creative" in 4th down situations is BALLS. Big, fat coconuts hanging between your legs that the players all get to slam with a sledgehammer on their way out of the locker room> And when you don't even flinch, they freaking KNOW that there's no 4th and 2 that they won't be asked to get.
BALLS.
Comments