First of all RT didn't throw a TD pass. I just looked at the tape again and that's probably down at the 1 yard line. You're right about the difference in distance. That would weight the value of that play more towards the RT INT using expected points added. But win probability is still heavily weighted towards time remaining so I'm still predicting the Watson INT was more important.. can't tell with certainty until we actually see that stat (can't find it right now for this game).
No I don't. You said they were "anomalies" or "inconclusive". You didn't say they were "not property adjusted". The stats aren't anomalies or inconclusive per se. It's that they weren't adjusted properly for era (for example).
I disagree. The hit dislodged the ball and the receiver went down trying to recover it. Had he properly secured it, he recovers/absorbs that hit and scores. In any case that was a perfectly executed play by Ryan Tannehill. The ball placement could not have been any better. That should have been a TD.
No that's an error in logic. A stat is neither "conclusive" or "inconclusive" on its own. Those words apply to arguments using stats, and adjusting stats for era won't change that. For example, saying Tannehill is ranked X or Y relative to others using either adjusted or unadjusted stats is just a true statement and is neither "conclusive" nor "inconclusive" per se. Wrong word to apply here IMO.
Tannehill was definitely not at fault I agree. And whether that pass would have been a TD or not, it's highly likely the Titans get a TD there.
Tannehill over Watson today! Let's see how he does at Houston. But according to my eyes, stats and fantasy.. I don't see how anyone can say he wasn't the better QB today.
https://www.espn.com/blog/tennessee...texans-plunge-dagger-in-titans-division-hopes Pivotal play: Ryan Tannehill found Anthony Firkser just short of the end zone, but Texans safety Justin Reid knocked the ball loose and linebacker Whitney Mercilus gathered the deflected ball for an interception. Mercilus returned it 86 yards before being taken down by Jonnu Smith. Watson found Kenny Stills on the next play for the first of their two touchdown connections. That and the kicker having the FG blocked were game changing events. Those were why the Titans were scoreless in the first half. But yeah, Tannehill wasn't "exceptional" today. Lol...
It looks like the INT thrown by Tannehill at the goal line was associated with a 131.7% change in win probability. The Titans were 70% likely to win before the play, and the Texans were 61.7% likely to win after the play. However, that certainly wouldn't be the case for every interception thrown relatively early in the second quarter of a scoreless game. What made the interception so impactful in that scenario was that 1) the Titans were about to score at the time, and field position is certainly factored into win probability, and 2) the interception was returned to the opponent's 12 yard line, making the Texans very likely to score at that point. The extreme change in field position was certainly more impactful in terms of win probability than the interception. A fumble returned in the same manner would've been just as impactful, and an interception thrown at the 50 yard line at the time, with no return, would've probably been almost meaningless in terms of win probability by contrast.
Ah! Found the link: https://www.pro-football-reference.com/play-index/win_prob.cgi OK, my intuition is wrong about the change in win probability, KeyFin is right about this. That huge difference in yardage is overcoming the difference in timing of the play. Using play-by-play data: https://www.espn.com/nfl/playbyplay?gameId=401128068 The Titans were 70.7% likely to win before the INT and the Texans 66.7% likely to win after it. So from Tennessee's point of view it went from 29.3% win probability to 66.7% for a difference of 37.4%. Before Watson's INT the win probability for Houston was 96.3% and after it the win probability for Tennessee was 16.5%. So from Tennessee's point of view they went from 3.7% to 16.5% for a difference of 12.8%. OK, I now agree that the play with most impact on win probability was the Tannehill INT. I guess I was right however about how difficult it is for humans to intuit this (in this case me). At least now I know that link so that should help with the intuition over time.
Yeah, the Titans "should have" been up 10-7 going into the half instead of down 14-0. If the rest played out the same, it's a 31-17 beating by the Titans. Two really tough breaks for them today...they were the better team.
Not only was it a massive swing in points and momentum, all other things being equal (which there's no way we'll ever know), that one play took Tannehill from a 116 passer rating (roughly his average) down to 92.2...so essentially he had the same game he's been having for the past 7 weeks, on average, including a clutch 4th quarter drive, unfortunately to no avail in the end.
Ok if it hurts some of your feelings deal with it but Jesus God Mary and Joseph... Some of you are starting to sound like a bunch of Sheldon Cooper geeks wannabes with all of your mathematical formulas and statistical probabilities. Football is a game played by MEN, not some stupid computer CPU and there are all types of influences on each and every play. Discipline, motivation, fatigue, strength, speed, agility, heart, desire...none of your statistical probabilities nor mathematical formulas account for any of these things. There’s an adage pertaining to football that been around forever...”anybody can beat anybody on any given Sunday”. Those of us who actually KNOW football knew this was going to be a hard FOUGHT game. Credit Houston for fighting and wanting it more than Tennessee. And observe I used the TEAM rather than individual players as football is, as I’ve stated until I’ve been blue in the face, a TEAM sport
Exactly. This is a living breathing relevant and recent example of the old fin d/resnor axiom: "It takes two people to complete a pass", and how that can translate to a whole performance and people's incorrect perceptions. With the Dolphins, Thill lived in the 90 rating area, was accused of not winning games and "choking" by not having 4th quarter comebacks and losing big games. This last game is a perfect example, of one player not Thill, making a SINGLE mistake and it caused: - Thill's rating to drop from a 116 to a 92 - Cost the game - Cost Thill a 4th quarter comeback - People will use the result of this pivotal game when the specifics of what happened are forgotten, to say Thill didn't come through in a big game Again, one mistake from another player. Imagine if many of the players you've played with over the years, weren't very good. Is it really a ridiculous notion that worse players make more mistakes? Now imagine, how that must change your approach to your position over time. Imagine the confidence that removes from you. The other issue is that most people think a good QB can get a team about 3 or so wins based on their talent or the not real thing "it factor". The funny part is, they only think that IF the team ended up over .500. They look at a QB with a record of 10-6 and say that QB's talent kept that team from being 7-9. But you never here them say a 6-10 QB kept that team from being 3-13....even though that happens.
And the other narrative is that Watson out-played Tannehill...which is complete garbage, they don't go head-to-head or impact each other's play one iota. However, even to make this comparison singularly is a joke. Is it not much worse to throw 2 INTs in the opposing team's endzone? Is this not where the rubber meets the road? Historically, there are tons of examples of a QB passing up and down the field all day, just to fail at getting in the endzone...those games aren't held up as an example of great QB play. Tannehill's "INT" was the result of planets being aligned against the play...just one of those "shake your head" moments. Watson's INTs, well... In the end, they both played well enough to win the game...but only one could.
If Dan Marino for example had high levels of focus, motivation, drive, grit, and heart, that was certainly very likely represented by some percentage of his his career passer rating. There is no reason why something numerical cannot represent something non-physical. Every person here has an IQ. That is an everyday non-physical thing that is represented by a number.
We need a 47 page thread with people bickering about Jordan Phillips and Kenyan Drake, ad why or why not they were good here in Miami. Make Mains Great Again.
If so, then their statements would be consistent with those of any scholar of human nature and its measurement.
Very interesting. FWIW This type of thing is why I like to challenge people and see others (Keyfin in this case) do it as well. It isn't about being wrong or right but rather about trying to look at things from all angles in order to determine what the truth of the scenario is. No matter how good someone is with their eyes or with statistics we all overlook things.
Sheldon Cooper is fake. I'm real at least Absolutely true. The more important question is whether there is a better approach. I personally don't see the point in criticizing an approach if you can't show how to do it better. So the question is this: can you tell us what level of "discipline, motivation, fatigue, strength, agility, heart, desire" any player in ANY game every played had? (I deliberately left out "speed" since that can be observed from tape) No right? So what's the point in pointing out statistical analysis can't do it when no one can?
Have your takes on CBrad, but I don’t think I’ve seen anyone in the game better at what he does.. Dude should be working for a team.
In a bottom line nutshell, because statistical analysis doesn’t lead men...men lead men and all of those intangibles are the result of men, either self motivation or motivation instilled by leaders That’s what makes winners, not stats
You didn't answer the question. The question is: what is the level of "discipline, motivation, fatigue, strength, agility, heart, desire" that any player in any game had? You said stats can't tell you that. I agreed, but I pointed out neither you nor anyone else can tell us that either. It's a pointless point to make if you're saying no approach we have as observers of the game will allow us to quantify every variable that matters. Of course you only singled out stats, which you shouldn't have done because watching tape won't tell you stuff like the motivation of a player or how fatigued he is either.
Regardless of whether or not we can pinpoint those stats, it's just as erroneous to assume that they either don't exist, or that they're the same for all QBs.
See, the disconnect here is that you seem to believe if it cant be quantified it isnt important. That seems like an extreme view, unless you want to argue the final statistics actually do capture those aspects of play because though intangible they are still accounted for in the play, and thus statistics, of players. That I would be a little more apt to agree with.
?? Guys.. I explicitly agreed TWICE that stats can't tell you what the state of all variables that matter are. You guys need to read the posts more carefully. For example.. in post #1867 in my first response to TDK I completely agreed with his argument saying "Absolutely true" but then pointed out what was more important, namely whether there's a better approach. In the next post #1870 I said: "It's a pointless point to make if you're saying no approach we have as observers of the game will allow us to quantify every variable that matters." Seriously, why would I say that kind of stuff if I thought they did NOT matter? Man...
Because usually you treat those things as all being equal for all QBs since we can't pinpoint what the actual values are.
I also always point out they matter but that we can't quantify them. So in the future don't accuse me of thinking they're not important. My view is you do the best you can even if it's imperfect. If you don't take that view you'll never get anything done.
For the record I'm not accusing you of anything. I said "it seems" which whether you agree or not, it does. If it didnt, a majority of people wouldnt be coming to the exact same conclusion. Also, people are merely pointing out you can see things that cant be put into statistics. No more, no less.
Well do me the favor of remembering my responses here. I’ve been saying the same thing for years so none of this should be new. And the problem isn’t pointing out an issue with stats. The problem is people being biased by only pointing it out for stats when every other method has the same issue!
I did not accuse you of thinking they aren't important. I said that you usually treat them as being equal for all QBs because we can't identify actual values for them.
Sure you did. Not in your original post.. but when I asked why I would post what I did if I didn’t think variables like motivation mattered, you tried to explain why in your response.
What are you babbling about? I've watched you treat oline play as the same for all QBs because you say that it all comes out in the averages over time. Quote the post you're taking about please.