Wednesday, October 5, 2011

Analyzing Richt's Recent 4th Down Strategy

There have certainly been some polarizing opinions concerning Mark Richt's questionable decision-making on 4th down. Specifically, I am referring to two recent instances:

  1. Electing to attempt a field goad on 4th and 1 at the Ole Miss 30 yard line, early in the fourth quarter with a 24-13 lead. 
  2. Also electing to attempt a field goal against Miss. State on 4th and 1 at the MSU 5 yard line, early in the third quarter with a 21-3 lead. 

Sure, we can make arguments for AND against the decisions; in the Ole Miss game, it would have given us a full two touchdown (14 pt) lead, and in the MSU game, we would have been up 21 points, or three touchdowns. As you may have noticed by now, I am using phrases like "would have" and "attempts", which is probably why most people are so upset over the decisions. But was Richt really in the wrong when he decided to kick these? Could the fans actually be the dummies?

First, let's examine -- objectively -- what the percentages actually were in each of the given situations:

1. THE OLE MISS DECISION -- Last season, the popular Michigan-centric blog, mgoblog.com, did a nice breakdown of 4th down conversion rates in college football. Here is the first usable piece of evidence I found: On average, college football teams successfully convert 4th and 1 attempts 72% of the time. This does not include kicking a field goal, it just means "going for it" and converting. Here is the second valuable piece of information: Teams that start drives at their opponent's 30 yard line score a touchdown 53% of the time (I imagine this could include any 1st and 10 situation on the 30, and not just the first play of the drive). And here is the last piece of information: From this distance, which was a 48 yard field goal, Blair Walsh had about an 80% chance of making it based on his career history (it should be noted that on average, most kickers are around 50% from this distance, so Walsh's limited sample size from 40-49 yards may have bumped his average a little high). Now, let's put these numbers into action:

  • If teams convert 4th-and-1 attempts 72% of the time, and score TDs 53% of the time starting drives from the 30 yard line, then essentially, we had a "72% chance of GETTING TO the 53%" (assuming we got exactly the necessary yardage on 4th down, which would have put us at the 30). So, if we multiply these number together, we get: (.72) x (.53) = .3816. Another way of saying it is we had about a 38% chance of scoring 7 points IF we had gone for it on that play (it doesn't mean we had to score on that exact play, but on the same drive).  Thus, to find our "point expectancy", we simply take 38% of 7 points: (.3816) x (7) = 2.67 points. Remember that 2.67. Now, if Walsh had an 80% chance of making the 3-point field goal, his "point expectancy" would be: (.80) x (3) = 2.4 points. Only 2.4 points vs. 2.67, and this is giving Walsh a staggering 80% chance of making a 48-yard FG, which is probably a little high. 
My final conclusion on this play: Richt was wrong, he should have "gone for it" given the percentages.

2. THE MISSISSIPPI STATE DECISION -- Football Outsiders has a nice chart detailing the chances of scoring at a given starting field position. What are the chances of scoring at TD from first and goal at the opponent's 5 yard line? About 85%. Thus, our "point expectancy" would be: (.85) x (7) = 5.95 points. But remember, we had a 4th and 1, which is only converted 72% of the time so...our point expectancy then would be: 5.95 x (.72) = 4.28 points. Even if you give our field goal team a 100% chance of making the 22-yard field goal (which didn't happen as you may have noticed), that still would only be worth an even 3 points, which is far less than the 4.28. 

My final conclusion on the play: This MSU scenario was an even worse decision by Richt, because in addition to the point expectancy difference, if we get stopped on the 5 yard line, MSU has to go 95 yards for a touchdown, which is unlikely to happen. Really strange decision making in my opinion. 

Final Thoughts: I realize Richt probably had ulterior motives, which did not include playing the percentages (to send a message to his team and coaches, etc.) and his general pissed-offness probably played a factor, but we have seen these funky decisions by him over the years. When it comes to fourth downs, Richt is the bizarro Les Miles. 


13 comments:

  1. Statistics!? Keep the statistics for baseball. In football we make decisions based on our gut / guessing. Just because statistics are critically important in baseball does not mean they could help in football :)

    ReplyDelete
  2. I like your work here. I agree 100%. Another smart guy from the wall street Journal did the same analysis years ago and found the same conclusion. Now, in Richt's defense, going up by two td lead is huge and should play into the decision; however, converting a first down and being able to milk another two minutes off the clock even if we don't score or have to kick a field goal anyway cannot be overlooked either. I think Richt's been burnt too much with his rare-aggressive playcalling. It seems it rarely works for him. On-side kicks v. UF and SCAR come to mind.

    ReplyDelete
  3. Wait a minute... you didn't factor in the points teams get from kicking a field goal after they convert 4th and 1, did you? It skews the favor even more toward going for it. Here's the deal: if UGA converted on 4th and 1 but didn't score a td, they would still have a chance to kick a field goal. We need a third piece of information to do the math: the FG success rate for teams that start a drive on their opponents 30.

    ReplyDelete
  4. There was also an instance earlier in the Miss St. game where we had 3rd and inches and didn't covert. I'm with Richt here. If you can't convert on third and INCHES why the hell would I expect you to convert on 4th and 1?

    ReplyDelete
  5. anon 10:12,

    Are you saying I should be writing for the Wall Street Journal? Ha. I agree totally with all of your thoughts and here is something I didn't point out in the article: if we had gone on to score a TD in both examples, we would have been up an extra POSSESSION (i.e. three scores vs. just two in the Ole Miss example).

    Anon 10:27,

    You're right, I didn't factor in the possibility of kicking a field goal LATER in the drive, which would have further supported the decision to go for it. As for the FG success from the 30, that info wasn't available but I bet it's pretty high.

    Anon 10:46,

    That's a great counter-argument to the "go for it" thinking, and I think Richt points it out in his post-game comments.

    ReplyDelete
  6. you shits will nit pick anything to death that doesn't work. If they'd hit those 2 field goals we wouldn't even be having this conversation. Frankly it wasn't even a questionable call, it was absolutely the ONLY call. Richt was right in that they were just stopped on 3rd and INCHES in one case, if you want the first down... want it bad enough to get it then. In both cases it was to put us up by a touchdown increment and given Walsh's history there were no thoughts that neither would be missed... let alone either. Add to that, if he is struggling we need to get some good kicks back under him in real games--at some point we're going to need him to win one.

    Let's wander aloud why those brilliant wall street numbers analysts don't coach football... you ready for it? Because this one is glaring...

    Given your percentages... with such a high success rate and scoring this way v/s that... WHY WOULD WE EVEN NEED A FIELD GOAL KICKER? WE SHOULD JUST GO FOR IT ON 4TH AND 99... SEEING AS IF WE'D SUCCEED 72% OF THE TIME!!!

    Numbers and calculators don't belong in football. Just scoreboards and guts. Using a calculator to decide to go for a field goal is akin to trying to convince a girl your willy isn't little, "if you measure the circumference I'm in the top 5% of all men".

    ReplyDelete
  7. Hal,

    First off, don't yell. It causes bad things like added stress and hypertension. It also irritates the people in the cave around you.

    Secondly, listen to the premise of your argument, "Numbers and calculators don't belong in football. Just scoreboards and guts." Can you make it on 50% of what "belongs in football"?

    We also enjoy nitpicking things that do work. Please see the post below for evidence (ie Malcolm Mitchell).

    ReplyDelete
  8. Hal,

    You can give Walsh "confidence kicks" in a blowout game, not in a relatively close game. Also, the 72% is on 4th and 1 situations only, not ALL 4th down plays (and for the record, I think teams SHOULD go for it on 4th down more). As for the "numbers in football" argument, Homer Smith is one of the most respected coaches in football history, and he was notorious for using statistical analysis when making situational coaching decisions. Besides, is kicking a field goal on 4th and 1 a "gutsy" football decision?

    ReplyDelete
  9. @martin--actually that wasn't a yell, but it was an emphatic laugh. Certainly not concerned about irritating anyone as irritated as I get reading such senseless babble. @dawgola--these weren't necessarily blowouts in terms of points on the board, but your argument contradicts itself? Were these close games where points would ultimately matter? Or were they blowouts? In neither case when the kicks in question were attempted was the outcome of the game in question. But, to your point, if they were close then we needed the points for separation particularly when you consider both kicks would have made touchdown increments and separated the game by 2 and 3 possessions respectively . If they weren't close then the confidence kicks were ok. Either way, I'm right.

    Having said that, the fan in me, wanted to go for it too from my seat in section 126. But that was an emotional reaction and a dumb one. The Great Homer Smith did use analysis and statistics to aid in decisions--all coaches do. Some better than others. A couple years ago Belima @ Wisconsin used those stats to go for two when he was up by 100 points or so... good call or bad?

    You can't let stats and wall street numbers analysts (they aren't even good at their own jobs right now) tell you how to coach football. Use them as guidance? Sure, why not.

    Several years ago, Florida lost a damn good football coach because their fans whined and bitched even when they won games, fugettaboutit when they lost. We used to consider ourselves better "fans" than UF fans based on things like that... can't say that's true anymore.

    ReplyDelete
  10. Hal,

    "In neither case when the kicks in question were attempted was the outcome of the game in question."

    The Ole Miss game was an 11 point lead at the very beginning of the 4th quarter, so yes, it was still in question . And the Miss State game was still in the 3rd (otherwise I would not bother going through these scenarios in a blowout situation) Also, in both cases, the game was 2 and 3 possessions respectively, and an added field goal would not have changed that (though it would have made them full multiple touchdown leads). A potential TD would have put both games away.

    BTW, I remember the Belima thing, and it was just ridiculous, as you have to use some common sense when sportsmanship is in question.

    ReplyDelete
  11. ok, tomato toma(ah)to... you win I win. In question--theoretically maybe. Possibilities get even better if pigs could fly. When a team needs red herring luck via trick plays and a bone headed pic-6 to even get on the board... game ain't in question. But if it really was in question, and they really did have an offense that could challenge us (which the did not), then points would be at a premium one would think. I remember once the great Bill Parcells kicked a short field goal, and on the play his kicker was roughed and they would have gotten a first down inside the 15. He declined the penalty, said you always take the points. They were no more a threat to us as we were to Boise State but I digress...

    potential TD's and close calls only count in horse-shoes and hand grenades. Any coach worth his paycheck would agree with exactly what Mark did on Saturday... knowing full well that armchair quarterbacks like you and me with a false perception of their own football acumen would begin the second guessing immediately. But that's why they're in the arena and we aren't... because they're better at it than we would be.

    Are the guys like Les Miles more fun to watch sometimes? Maybe so. I still doubt Les would have done what you're asking in this circumstance, he'd taken the points too. Now if he were down, or it were a truly close game... different story. But in that case Mark would have too... he's gone for it plenty to fair success. Also when Miles' name comes up I can't help but remember just a few short years back, LSU fans wanted his tail. They wanted him gone. "No one does Les with more" they said. Fans are fickle and frankly quite dumb sometimes. Makes it hard to just enjoy Saturday afternoons in the most beautiful place in all the land. But I do, and I'll be blowing up K'ville in 3 days chasing my team.

    ReplyDelete
  12. Hal,

    Don't get me wrong, I am NOT saying I want Les Miles, or that he is a better coach than Richt, just that he is typically more prone to make surprisingly aggressive calls on 4th down. And I agree with you, DURING the game, as a fan, I am not thinking about these percentages, I am just rooting for the team (but I still thought -- during the game -- we should have gone for it on those fourth down calls ;)

    ReplyDelete
  13. Hal,
    You fear what you do not understand. Many good coaches may have kicked a FG in those situations. Some would not. Kickers do miss FGs. It's silly not to consider that three points is not a certainty; and tell me, is it time to take thought that Walsh may miss?

    http://sports.espn.go.com/espnmag/story?id=3641375

    To be fair, I share this criticism for coaches at large. There are some coaches who've embraced this notion somewhat, one coach in particular that I hate but must say has figured this part out.

    ReplyDelete