Second Guessing Curious Decisions In The Stanford Game
by Brian Floyd on Oct 27, 2010 2:02 PM PDT
Email
PrintI don't like playing revisionist with football decisions because hindsight is almost always 20/20. When a decision fails, it's easy to look back, decide the opposite call would've worked and assume it was a poor decision. However, there were two decisions head coach Paul Wulff made against Stanford that stood out to me, not because they failed, but because they defied what the odds say a coach should do.
After the game, Christian Caple, who does a great job covering WSU and everything else in the area for the Moscow-Pullman Daily News, sent out a few tweets that caught my eye.
Instead of praising WSU for not giving up, maybe ask why they didn't go for 2 on TD that cut it to 38-21. Conversion, a two-score game...
Christian Caple
ChristianCaple
...Follow with an onside kick, clearly still trying to win. Why not REALLY try to win?Christian Caple
ChristianCaple
Was he right? Read on for that and more.
At the time, it didn't dawn on me that going for two was the right play. It didn't even cross my mind until I saw what Caple wrote. So I looked into it, finding the odds and the two-point conversion charts many coaches keep in their back pocket. You can find that chart here.
The chart breaks puts a number on almost every possible situation a coach can go for two. The decimal numbers contained within the table can be interpreted to find when it's the right decision to go for it. Assuming the success rate for a two-point conversion is 40 percent, the rate given on that page, a team should always go for two when the cell is less than 40 percent. With three minutes to go, down 18, the no-brainer decision was to go for two.
The other decision -- or really the non-decision -- came in the second quarter. On third down, Jared Karstetter had enough for the first down but jumped back to avoid a defender, leaving WSU inches short of converting. With a fourth and inches from their own 36 and just over five minutes to go in the first half, the Cougs punted.
The line of thinking on fourth down calls is simple in terms of probability. No matter where a team is on the field, going for it on fourth and two or less is the right play. In fact, when 65 yards from the end zone, going for it on fourth and four or less is the right play. The conversion success rate for a fourth and inches from outside the opponents' 20 is around 75 percent. The odds are in the offense's favor.
Considering the WSU defense hadn't been able to stop Stanford in the first half -- outside of a Kevin Kooyman interception with Stanford deep in WSU territory -- there was no reason not go for it. Instead, Reid Forrest uncorked a booming kick into the end zone and Stanford drove 80 yards for the score -- putting the Cardinal up 24-7 going into the half.
We don't know if these two decisions would've changed the course of the game, and it doesn't really matter to me. What does matter is electing to kick the extra point and punt on fourth and inches flew in the face of some overwhelming odds at the time.
I highly suggest wading through the slides on fourth down odds and checking out the two point conversion table if you want to know more about the math behind some of the decisions coaches make throughout a game. The formulas for fourth down conversions are much more complicated than I outlined here. To get an idea of how the conclusions are drawn, read more about them here.
32 comments
|
0 recs |
Do you like this story?
Comments
Onside kick...
Too risky against Stanford. We didn’t train for it so why take the risk of getting another score upon us.
by well you win some and lose others on Oct 27, 2010 2:12 PM PDT reply actions
There's nothing in this about an onside kick.
Caple was talking about the fourth quarter, when WSU did kick onside down 38-21.
Curious moves.
There was another one from the Stanford game. With about 1:35 left on the clock Stanford just got the ball down to the two yard line. They run for a one yard gain. I immediately said “TIMEOUT!”
Instead, WSU did not call a timeout, and Stanford let the play clock run down to 1, and then they called timeout. The next play, Stanford punches it in, and the cougs are left with 42 seconds on the clock. They run one play and went to halftime. Why didn’t Wulff call a timeout there to give us about 1:30 to try to score before the half? We have deep threats.
I don’t have a lot of respect for Wulff’s game/clock management decisions.
I didn't look at it because there's no way to tell what was going through his mind.
With one timeout left was there a reason he wanted to save it? Did he just want to get into the locker room and regroup? I have no idea nor do I know the circumstances surrounding it.
Yeah, it's easy for us to apply the video game mentality and think you always have a chance
I’d say most of the time, a TO is the right call, especially with how explosive our offense has been. But there might have been a legitimate reason, chief among them that if you do burn that timeout, and you do fail to get a first down, you might end up just giving it right back to them with more than a minute to go.
Stanford was out of timeouts
So I guess you could think if Stanford gets stuffed on second down, the clocks running, they have to hurry up and are forced to throw on third down to stop the clock in the event they don’t get in to kick a field goal.
Since Stanford was out of timeouts (hat tip Brian)
you can pass on first and second and if nothing happens run it on third and the clock should be about gone.
by spencer peaty on Oct 27, 2010 2:43 PM PDT up reply actions
Didn't realize that
That would seem to lend support to the idea that we should have used our timeouts.
I understand what your saying.
I would understand (although not agree with) the desire to just get into the locker room. But If he was wanting to save the timeout, that would be poor logic because any timeout you use on offense will not save half as much time.
I don’t have a problem with isolated issues like this, because coaches know their team better than we do. But it seems the list of questionable decisions is mounting. When we get to the point where we are in really close games in the final minutes, it could become an issue.
by spencer peaty on Oct 27, 2010 2:41 PM PDT up reply actions
I'm with you.
The clock management has been very questionable at times. Sometimes it feels like we are playing not to get blown out, instead of playing to win.
I think isntances like these are why people are questioning Wulff
Since we appear to no longer be getting blown out.
Some of these are just real head scratchers.
by B-Lot tailgater on Oct 27, 2010 4:10 PM PDT reply actions
Lots of my opinions of Paul Wulff have changed this year.
His game management is not something that I have changed my opinion on, however. Now, obviously, we haven’t been in many of these situations so he hasn’t had much practice. But his late game management just has not impressed me even a little bit this year.
There were horses and a man on fire and I killed a guy with a trident -- Brick Tamland
my thoughts
Say there are 250 problems (or things to fix) with this team. By NCAA rule, he can only have a certain size of staff and certain amount of time with his players. Potentially its a matter of taking inventory on what he needs to spend the bulk of his time working on, then prioritizing accordingly. It’s guaranteed that certain components are going to get left out while he tries to bring other elements up to speed. Not saying he doesn’t have a 2 pt conversion chart or can’t have one on hand to help him out.
That being said, when you look at a 75% probability that a team who goes for it on 4th and less than 2 – will convert… you do have to take it a step further. Most teams that go for it on 4th and short have 1 or 2 key elements going for them to feel confident enough to attempt the 4th down conversion to begin with: 1) they have been consistently running the ball well on that particular defense and 2) a great confidence that if they don’t convert then either their D is going to step up; or their offense is potent enough to keep up, score-for-score.
I think most of us would honestly agreee that 1) we haven’t established a run game all season long, 2) we have no confidence in our defense in making stops; or our offense being consistent enough to put up high 30s or 40s point totals to match the likes of a Stanford, Oregon and Arizona team.
Personally, I think he’s balancing-out keeping his team hungry and not heart-broken. He’ll throw out an onside kick and get his guys pumped up, but if a few of those type of plays go against us, he knows our players might be crushed. We already had 2 or 3 fumbles early in the game, I don’t know if he was ready to send one more negative shock-wave to our player’s egos just before the half.
by LeaveItToWeaver on Oct 27, 2010 5:46 PM PDT reply actions
On point 2,
Isn’t that exactly why we should be trying to keep the ball?
CougCenter In Reid We Trust
Twitter!
by Craig Powers on Oct 27, 2010 5:55 PM PDT up reply actions
we should always "try" and keep the ball ...
We have the first 3 downs of every possession to do that. ;)
With Stanford having one of the worst punt return teams in the nation and us having the Heisman – in that situation you have to feel good about pinning them deep and forcing them to go 80-90 yards for a score. On top of this Owusu was injured.
If our offense was “high-powered” (like an Oregon or Boise), then sure give them the ball on our 35 yard line all the time and we’ll just match touchdown for touchdown. The reality is we are improving and getting back to being an exciting offense to watch, but we are not in 2002-2004 WSU offensive form, just yet.
by LeaveItToWeaver on Oct 28, 2010 1:35 AM PDT up reply actions
The 75 percent probability doesn't have anything to do with those two items.
It doesn’t matter if a team establishes the run or has confidence. When looking strictly at the statistics, which take into account all different kinds of teams that have different offensive philosophies, the decision doesn’t even fall to 50/50 until you get to 4th and 5 outside the opponents 20.
I chose both of these situations because they’re absolute no-brainers from a statistic standpoint. The two-point conversion decision is such a no-brainer, the chart says if a coach has any chance — any at all — that it will be successful, do it. These are two very basic decisions that shouldn’t require that much thinking for a coach.
So what does not going for it do to his players psyche? Well, he’s now told the offense he doesn’t think they could get half a yard, even after they cut the deficit to 10-7 before Stanford scored again. He’s also told the defense he doesn’t trust them to keep Stanford out of the end zone. Neither one of those are confidence boosters.
when in rome...
I’m a research/numbers guy so I appreciate these type of pieces, but I just can’t get behind a “go for it every time you are in a 4th-and-1 situation,” especially with no run game what-so-ever. I don’t need probability to tell me that is a terrible decision for the Cougs.
http://mgoblog.com/diaries/fourth-down-and-worth-it
The blog link above discusses the alternative side to these statistical theories and points out some of the flaws. They basically say that by holding true to these theories you are potentially squeezing out 1 win per 4 years, but there are not clear probabilities for reverse over the course of 4 years either. So potentially you get an extra win every 4 years, but you might also acquire an extra 1 or 2 losses in that same 4 year span.
Going back to my original thought, these are all probability-based on past decisions, but these past decisions are not all based 100% on expected points analysis. We don’t know how many coaches are using strict numbers and probabilities, compared to coaches who are basing their decision on personnel/game-time situations/weather/clock management/etc. Sampling generally is a fine strategy as long as the data captures an accurate depiction of the entire audience. In something so small as 4th and 1 opportunities you are working with a small sample size and that sample size is going to be skewed towards teams that possess a physical advantage over the defense – enough so to give them the confidence to attempt it in the first place! With any type of research, you have to put your data/sampling together so you are eliminating bias. Going for it on 4th and 1 skews so heavily towards good offensive teams, you can’t get a clear reading on its probability for a team like ours.
by LeaveItToWeaver on Oct 28, 2010 2:18 AM PDT up reply actions
I'm not sure why going for it on 4th and 1 "skews so heavily towards good offensive teams"...
… as you provided no evidence of that and it is counter-intuitive. I think Brian has made a valid point.
The other aspect of this decision, even if you don’t expect to win that particular game, is getting your boys a “live ammo” rep in a high pressure situation. At some point this season, they are likely going to have to convert one or two of those to win a game. Having run it before would be useful.
Damn, my eyeball tastes good.
This comment really illustrates...
What my main critique of this staff would be so far. I think so many of their decisions are based on not putting the players in a position where they may fail, instead of just telling them what you expect of them and holding them accountable. I think this approach really hurt the “killer instinct” on this team, especially amongst the veteran players, who almost to a man, have regressed in their play since Wulff took over. The new comers seem to have a much more agressive mentality when they take the field, and that may be due to them feeling more secure in being Wulff’s guys. Early on the cautious approach of not running our no-huddle offense and our sit back and keep everything in fornt of you defense, were both predictated on extremely low expectations for what this team could do, and lo and behold, they met them. I have been happy lately to see that they have been removing those training wheels slowly, but I wonder if they did more damage than good.
by LongballWSUFB on Oct 28, 2010 9:30 AM PDT up reply actions
Great piece
I had noticed the 2 pointer and jokingly said to my Coug buddy as we watched- go for 2 make it a 2 score game. I was kidding… but it did strike me a logical from a ludicrously myopic standpoint. I said to him “A stop- a score 8 point lead after the two pointer. Onside bomb to tie it -as the clock runs out. Pissing off Harbaugh to no end” .
I don’t know about the 4th down play. They went 80 with relative ease. Do you give it to them with less yards to go? They score that much quicker in the silly 20/20 hind sight world of every drive going exactly as they did depsite the momentum chance of the the fantasy scenario.
I do feel we need to allow the staff a chance to regain the confidence that they too had battered out of them as they watched the horror of ’08 and ’09 unfold. Every game right now is a lesson… for the young ones and the staff.
People barked about the weird pooch kicks as dumb in the early games… yet I noticed in the SMU game that #4 (Carpenter?) missed a perfect onside lob kick by not looking for the ball. It then worked later on and the ‘weird kicks’ had meaning. Who knows if the SMU game is different if Carp snags that onside kick? But perhaps the coaches need to relearn certain aggression football techniques too.
Again great piece.
If you can't Go Cougs... don't go.
One HUGE problem with the "go for two" argument.
It’s a 40% shot to get the conversion each of the 3 TDs we’d need to tie, but we’d have to do it 3 times to tie in 3 scores. That drops our chances to (if my math is right 40% x 40% x 40% is) about a 6% shot at the tie. If you get TWO conversions (16% chance), but miss the third, you’ve still got to get the ball back a FOURTH time and score a FG to win. If you make ONE of the conversions but miss the other two (which is the MOST likely scenario, statistically), you now have to score four touchdowns to win.
All while limiting Stanford to three and outs without them running any time off the clock.
However,
I agree with your 4th and 1 argument. And BTW, bad, bad play by Karstetter to step back on that one.
by TiltingRight on Oct 28, 2010 11:09 AM PDT up reply actions
*four touchdowns total
or three more touchdowns…
by TiltingRight on Oct 28, 2010 11:10 AM PDT up reply actions
Why is it 0%?
Not sure what you’re saying.
My point is that of the four scenarios (getting 3 TDs and converting 3x, 2x, 1x or taking the PATs) the only one that doesn’t require a 4th possession in less than 4 minutes has a 6% shot of succeeding IF we manage to get the possessions to begin with, and the most likely scenario leaves us needing 3 TDs rather than 2 TDs and a FG.
by TiltingRight on Oct 28, 2010 11:11 PM PDT via mobile up reply actions
Comments For This Post Are Closed
Secondary Sidebar
User Tools
Community Guidelines
Related FanPosts on CougCenter
Related FanShots on CougCenter
FanPosts
Community blog posts and discussion.
