Exactly…completely overblown, much as the public puts too much into the loss of a star player, or HFA.
The math behind going for 2 when down by 14 late in game and scoring a TD.
- Thread starter Grant Green
- Start date
I'm pretty familiar with the basics (WOBA, WHIP, FIP...) but never did a deep dive. I tried building a baseball model years ago relying heavily on sabermetrics, but not something that I could keep up with while working a full time job with kids. I didn't know McCracken did soccer. I'll have to look into it.
Yea, I communicated with him a few times on different things. Really, really smart guy.
The DIPS stuff just blows peoples minds…that generally, pitchers control Walks, Strikeouts, and Home Runs. Beyond that, it’s almost all luck as to if a ball in play is a hit or an out in terms of what the pitcher can control.
Your addiction to sports wagering suggests that your naturally challenged musical career isn’t prospering
Last edited:
I like these stats better.
1 pt conversions have a success rate of 95%
2 pt conversions have a success rate of 47.5%
Based on this, if you're down 14 and you score 2 TDs, you're better off kicking the extra point because the odds show that you'll probably end up with 14 pts. Now if you decide to go for 2 after scoring 2 TDs, the odds say you'll be successful less than half the time which will leave you with less than 14 pts.
Kick it an get the 14 pts.
en.as.com
1 pt conversions have a success rate of 95%
2 pt conversions have a success rate of 47.5%
Based on this, if you're down 14 and you score 2 TDs, you're better off kicking the extra point because the odds show that you'll probably end up with 14 pts. Now if you decide to go for 2 after scoring 2 TDs, the odds say you'll be successful less than half the time which will leave you with less than 14 pts.
Kick it an get the 14 pts.
What is the two-point conversion success rate in the NFL? History and strategy behind the play
The two-point conversion is one of the more curious aspects of football. Here we take a look at what it’s all about and why teams are using it more and more
en.as.com
Last edited:
That is playing to lose according to the numbers and the objective is to win. You are trying to score 3 times instead of 2 using your route and most coaches are going the way of the math currently. Maybe as more choose to do this the data available will grow and the it may spit out a different trend. That is the thing with analytics as more data is gathered more informed decisions are made. Long story short I don't want any HC that completely ignores the numbers as they are basically saying they don't want to make informed decisions. Not that they have to follow the cheat sheet 100% of the time as there is some gut feelings involved on how the game is going, injuries, fatigue...etc. It's foolish to ignore the trends though just as it's not smart to just roll strictly with the numbers.I like these stats better.
1 pt conversions have a success rate of 95%
2 pt conversions have a success rate of 47.5%
Based on this, if you're down 14 and you score 2 TDs, you're better off kicking the extra point because the odds show that you'll probably end up with 14 pts. Now if you decide to go for 2 after scoring 2 TDs, the odds say you'll be successful less than half the time which will leave you with less than 14 pts.
Kick it an get the 14 pts.
What is the two-point conversion success rate in the NFL? History and strategy behind the play
The two-point conversion is one of the more curious aspects of football. Here we take a look at what it’s all about and why teams are using it more and more
Actually, those percentages would tie right back to Grant’s initial post where going for 2 would be obvious.
It's not obvious because if you kick the extra point 100 times and you go for 2 100 times, you'll end up with the same exact point total of 95, as long as the odds play out. There's no advantage to going for 2, however if you do and you aren't successful on the first attempt then you're forced to go for 2 on the 2nd attempt. If you take the extra point on the first attempt it gives you flexibility.
Did you read the initial post in this thread? There IS an advantage to going for 2 after the first one. Looking simply at expected points doesn't take into account the different permutations to getting to those expected points, and what they'd mean.
Based on your numbers, converting the initial two point conversion would give you a 95% chance of winning in regulation (a 45.1% chance of this scenario with the .475*.95). The odds of missing both to lose in regulation are 27.6%. Any other outcome puts you in OT (and again, this is all assuming that you're late enough in the game for it to all matter).
I just reread the initial post and I do not see a scenario where you kick on the 1st TD and then go for 2 on the 2nd TD. The OP discusses going for 1 on both or going for 2 on both.
Having said that, you'll never convince me that going for 2 on the 1st TD is the correct play. You should always go for 1 on the first TD which affords you the flexibility of choosing to go for the tie or the win on the 2nd TD.
There were a bunch of thoughts on the going for 2 after the 2nd TD in this thread (but it's obviously pretty long now), but here is the argument why you'd go for 2 after the first rather than the 2nd.
You go for 2 after the first TD and miss, and you still give yourself the ability to go for 2 after the 2nd TD to tie up the game. If you put all of your eggs in the "going for 2 after the 2nd TD" basket, you lose out on that other potential path to victory, as if you miss it, you're toast.
If you're going to miss your two point conversion, why wouldn't you want to still give yourself the chance to make up for it and still take the game to OT?
The math in the OP is simple. What is it that you don't believe?
Your odds decrease to 50% (actually less) by waiting for the second TD to go for 2 instead of the first.
I believe my gut, common sense and what I see. I can't recall a time when I've seen this work, in fact I've never seen it work for PSU.
Why don't you believe the math? Please point out if there is an error.
Green Bay and Tennessee titans both won games this year doing it. The tenn vs Miami game was an amazing finish. Green Bay made the playoffs by 1 game.
We've tried it once ever, so obviously the odds are that you haven't seen it successfully used by PSU...in general, coaches are just catching up to the math on it.
It worked twice in the NFL this season...
Packers 18-17 Saints (Sep 24, 2023) Final Score - ESPN
Game summary of the Green Bay Packers vs. New Orleans Saints NFL game, final score 18-17, from September 24, 2023 on ESPN.
www.espn.com
Titans 28-27 Dolphins (Dec 11, 2023) Final Score - ESPN
Game summary of the Tennessee Titans vs. Miami Dolphins NFL game, final score 28-27, from December 11, 2023 on ESPN.
www.espn.com
And on the baseball side, coaches believe in their "gut" or "common sense" forever and wouldn't be crazy enough to shift 3 infielders to one side of the infield. Then, a few coaches started doing it, everyone saw how effective it was, and MLB had to change the rules because it gave teams such an overwhelming advantage and hurt offenses.
Can you show me a comparison of how many times it worked this year and how many times it didn't?
Since 2018 in the NFL when this became a "thing", teams have gone for 2 on 23 occasions when down by 14.
14 times, they were successful on the try. 11 times it didn't factor into the game in the end (either the other team scored again, or the team that succeeded didn't score a second time). Once, a team was successful and ended up winning the game, but there were multiple scores so it didn't play as much of a role in the outcome as the intended scenario. And then those two times referenced in the prior post where this very much helped them to victory.
9 times, they were unsuccessful on the try, but never scored again or the opponent scored again, so it didn't determine the outcome.
And never has a team missed on the first try, and then scored again to have to attempt the second try.
But while these numbers would push even harder for the "go for 2" strategy, I'd put little stock in them since it's such a small sample size.
Thank you. I think another factor which needs consideration is the level of competition, which the percentages won't account for.
That's not really a good comparison unless you mean the times the 2 point was unsuccessful but the team then scored again to cut it to 2. A better comparison would be the number of times this strategy worked vs. the number of times the strategy of going for 1 in the same situation worked.
Not sure if you ever listen to Gill Alexander's show. That's who was discussing this and they Monty problem the other day. He and Steve Fezzik were talking about it early in the season and Fezzik (I know insufferable, but smart) thought that 80% of NFL coaches were now hip to this idea. Gill pushed back saying he thought it was 50% at best. I started tracking it through the season, but didn't follow through. Seems like it is still closer to 50%, but increasing steadily. I still can't believe Belichick didn't do it, but Todd Bowles did. Not a slight against Bill, he's usually on top of these kinds of things.
This was covered in great detail earlier in the thread. In theory, this is correct. However, assuming your team has a 50% chance to win in OT, your 2pt rate would have to drop below 38% to make going for 2 a worse option. And as Erial astutely pointed out, if you are that bad at short yardage, you probably aren't a good team and your odds of winning in OT are less than 50%, which changes the equation. It's all about comparing your 2pt conv rate to your chances in OT.
You lost me at Fezzik, lol. We used to post on some of the same forums back in the day and “smart but insufferable” is the perfect description. I haven’t ever listened to Gil’s show, but obviously it’s a topic that interests me.
I know. He's so arrogant that I hate to listen to him, but he's often right and I never turn down good info. I think you would like Gill (show is called A Numbers Game and available on Iheartradio). Big analytics proponent and his personality is the opposite of Fezzik.
Great explanation but they are just numbers. IMHO, going for two and missing is a problem for the mo's: morale and momentum. If you score that first TD and kick the XP, you have great mo (95%) but if you go for two and fail, you've lost a lot of the mo (about 50% of the time).
In addition, I think the success rate for XPs is much higher when it counts as many of the missed ones are when the effort is in the middle of the game and concentration wanes. I also have to wonder what the 2 P percentages are in crunch time versus the middle of the game when a team is trying to make up for a missed XP.
They are just numbers? But you can really break down mo after just scoring a TD with the "mo meter" out there on the market. I'd say I'm surprised by these takes that using data is bad or just not as good as a gut feeling, but I'm not really to be honest. These NFL and college coaches with PAID people dedicated to this stuff really need to trust the guy on the couch or in the stands....seems reasonable. Data science or Jim Bob's gut call? As data is collected since this is a relatively new should they just ignore data either way or use it to make informed decisions? Maybe the staffs could stop breaking down tapes too as they can just get a gut feel for it as using past data to help you really isn't as good as the old gut feeling. Football has always been about finding trends and using them to get an advantage to win, but change or progress scares some people.
Last edited:
Regarding XP percent, I feel very confident that the % will go down late in the game in a situation where your kicker needs to tie the game to go to OT vs in the middle of the game where the pressure is lower.
Agree. Fezzik just really rubs me the wrong way...I'll give him all the credit in the world for his back-to-back Supercontest, but he's really milked that for all he could. The past week, he started a twitter war with Ed Golden, the guy that runs RAS, who is as solid as it comes in this industry (I used to work closely with him)...Fezzik pretends he's this bastion of integrity when he went full scamdicapper about a decade ago. And now that I think about it, I have listened to Gill's show before when Ed was on it.
Great points but data doesn't always apply to humans. While there is certainly science to the game, there is also art. An example is salesforce.com. Salesforce, or something similar, is used to plot the sales process from start to finish so that executives can measure sales and plan for volume. But every good sales person knows that the buying process is an emotional journey. Knowing when to slow down, speed up, cajole, ridicule, call BS, coach up, are issues of art. They cannot be forecast or taught.
In my early days as a salesperson, I was in my hotel room when my phone range at 10:15 pm. It was a senior executive I had met only once. And he said "Obliviax, we are considering recommending your product tomorrow. However, I wanted to call tonight and ask you one thing. Are you going to take care of me? I am putting my career on the line and I have a non-working wife and three young children. When the **** hits the fan, will you take care of me and my company?"
Yikes.
Technology is great and there is no more highly supportive person about technology. However, technology is a baseline. Sure, early adopters do great but sooner or later, technology is a baseline. The differentiator is not just the early adopter of the science, but who can also manipulate the art. A friend used to say that one day, someone invented the shovel. And early adopters to using the shovel progressed in society. But sooner or later, everyone had a shovel. At that point, it went back to the human art of being able to envision what can be done with the shovel and not the shovel itself.
The data will forever be evolving and you should use it when clear trends or decision points state it is probably in your favor. It's like owning a casino and lowering your odds because you simply like the game better a certain way. It doesn't make much sense. You are then comparing sales to a coaching decision which again makes this about you and not a football coach making a decision based on the most current and useful data out there. Emotional decisions can work sometimes, but it's a fools game in the end to ignore data when it is out there for you.
This isn't about Obli's decision as he isn't a coach. This is about a head coaches using the most current and relevant information they have available to them. To ignore it would be ignorant, but they shouldn't treat it like a religion either with blind faith. There is a balance, but I'm going with the data scientist in the sports field over the salesman up in the stands.....and that is what most coaches will do to. Just like they all went that way in baseball and other sports. You cannot take the human element out of any sport such as dropped passes, missed blocks, or even players just tripping, but you can find certain things that work in your favor and then use those things to make more sound decisions.
This thread demonstrates two of my list of inherent human foibles. Reasonably intelligent people struggle with probabilistic thinking and once a person makes up their mind about something, it is nearly impossible to change it, even when presented with reliable data/logic.
It's odd as nobody is stating you have to use the cheat sheets 100% of the time, but you would be an absolute fool to ignore them as well. These numbers will change I think in time as the sample sizes grow and do you want a coach who can adjust on the fly to those trends or one that says....well back in my day this is how we did it...so that's that!!!! IMO, there isn't a competent AD or NFL owner around that would hire the guy who argues against analytics or data science in 2024. Maybe the Jets are run this way? Sorry to any Jet's fans.
Fair enough but people aren't rocks or engines. People choke. They are emotional. they make non-rational decisions. Casinos are a totally different story. That is simply math. Emotion doesn't change the cards unless you have people playing against people (poker, for example).
All coaches have access to the probabilities. That isn't a differentiator. The best coaches read the players. Do the numbers discuss the probabilities on the road? Do the probabilities discuss the fact that your WRs are gassed having driven the field (or the CBs)? Do the probabilities take into consideration that the LT turned his ankle celebrating the TD and isn't 100% for the 2-point try? Does it take into consideration the opponent's red zone competency on defense (or yours on offense)? All of these have great significance beyond the numbers. There is a reason why KC excels in tight games while the Bills and Cowboys do not and it isn't reflected on the Excel spreadsheet.
Bill Simmons used to joke that owners should take NFL coaching candidates to a blackjack table before they hire them (he said that after playing at the same table as Norv Turner one night when Norv kept sticking on 16 again and again and refusing to double down in +EV scenarios).It's odd as nobody is stating you have to use the cheat sheets 100% of the time, but you would be an absolute fool to ignore them as well. These numbers will change I think in time as the sample sizes grow and do you want a coach who can adjust on the fly to those trends or one that says....well back in my day this is how we did it...so that's that!!!! IMO, there isn't a competent AD or NFL owner around that would hire the guy who argues against analytics or data science in 2024. Maybe the Jets are run this way? Sorry to any Jet's fans.
As I've said a number of times throughout this thread, any variance in the probabilities due to human or other factors will not be enough to overcome the ~12.5% edge that going for 2 gives you. I don't think people realize what an enormous edge this is in a football game. And going back to the human concept of momentum that everyone wants to focus on, all of the attention is to a team losing momentum if the first 2pt conv is missed. If you believe that momentum is a factor, don't you gain momentum if you make it (which will happen 50% of the time in our given scenario)?
I disagree with your basic premise here:
Option 2: Go for 2 on the first TD. There is a 50% chance that you win the game right then and there (remember, we assumed that you will score again and kick an XP).
If you miss it, you still have a 50% chance to make the 2Pt conv on the second TD and then a 50% chance to win in OT. The probably of this scenario playing out is 50% (that you miss the 1st 2pt conv) X 50% (that you make the 2nd 2pt conv) X 50% (that you win in OT) = 12.5%. The chance of making two on your first TD is just less than 50% so you will end up being either behind by 6 or behind by 8. If you score a second TD, you not only have already run your best 2 point play, but you have less of a probability .50 x .50 is only a 25% chance of success. But by that point, you have exhausted your options. if you missed the first chance, you MUST go for two on your second TD.
I submit that you are better off to kick the PAT, all other things being equal, on your first TD. You are almost guaranteed of having momentum AND knowing that a TD puts you in a position to win or tie the game. So you score the second TD. If you missed the PAT you must go for two to tie the game. If you made the first PAT, you now have a choice of tieing the game or trying to go ahead in the game by either kicking the PAT for going for two. I like having the choice. Why? You don't know when you will score the second TD and how much time is on the clock. That makes a difference to me. Second, you don't know what the game situation will be when you score the second TD. Is there QB healthy? CB? S? LB? Best pass rusher? Best run stopper? Is their sideline melting down and in disarray? These things, you have no way of knowing.
I'd kick the PAT after the first TD and have all of my cards, depending on game situation, when I score that second TD. I can always go for the tie or win on the second TD and didn't eliminate any options.
The bottom line is, I agree with the pure math of it but games aren't won with math.
You eliminated a huge option here...what to do in the ~50% of the time that you miss your two point conversion. If you wait on it, you're screwed. You miss it after the first TD, and you can still make it after the 2nd TD to go to OT.
I see the relationship but i disagree that your 2pt success rate has to be the same as your OT success rate. There are definitely some offenses who would fare comparatively better in 25 yards of OT competition than 3yards of 2pt conversion. Washington's offense comes to mind. Others would have better numbers with the 3yard push.
And of course the kicking games and defenses, as we've covered.
It's fair to say that kicking an extra point first gives you the flexibility on your second score, but also means there are scenarios where you kicked it when you really needed to be aggressive: scenarios where the second 2pt fails, that first XP didn't help you.
Great thread.
I would suggest you read the OP again because you have missed the point. I'll try to simplify.
If you kick the XP on both TDs and go to OT, probability of winning = 50%
If you go for 2 on the first TD, your odds of winning the game = 50%. However, the benefit to the second option is that even if you miss it, you still have a chance to win the game by getting the 2nd 2pt conv and winning in OT. Now your odds are greater than 50% - in fact they are ~62.5%.
If you go for the XP on the first TD and decided to go for 2 to win it on the 2nd TD is 50%. Same probability as if you just kicked the XP on both TDs.
Don't believe me? Here is an ESPN article explaining it. https://www.espn.com/nfl/story/_/id...y-nfl-teams-keep-doing-why-analytics-backs-up
I don't agree. If you kick the XP after the first one you are down by 7. That is almost automatic. The chances of you missing it are less than 5%. So you score your second, now you can kick to go into OT or go for the 2 pointer to win it outright. You've eliminated nothing.
