In partnership with CBSSports.com
Online Now 783
Online now 454 Record: 11761 (2/27/2012)
The Web's No. 1 forum for coverage and discussion of Terps sports
Visitor discussion of University of Maryland and college sports
A place for lively discussion for all other sports unrelated to Maryland athletics
Feedback for IMS and 247Sports
You have no favorite boards.
The most viewed topics.
The most replied to topics.
The most up-voted topics.
The most down-voted topics.
The most up-voted posters.
The most down-voted posters.
The most followed posters.
Just try to think of it in terms of poker where the better players will be able to get in that 80/20 situation more often and whether or not the shot falls is like the runout of the cards.
Yea I totally agree with that and that is what I was TRYING to say at first. "Luck" is a factor but that doesn't make close games complete coin flips based on random chance. Unless you run like Adam Cook and consider the being behind in 80/20s flips.
Of course I played sports. Have you taken a probability or stats class?
I didn't say the difference is meaningless - and I think you're overlooking the definition of "close games." It's not games that are close with less than 5 or 2 min to play, it's games that finish within a certain range. Teams that crap the bed get close and then lose by 10 on late FTs, but games that finish close are essentially random. By definition, you've filtered out the games where one team has successfully capitalised on whatever intangible advantage they may have.
Also, the real issue is that you can't separate signal from noise with a small sample size. Having 6-10 games finish in a season within a possession or two is not a large enough sample size to have any hope of identifying who's "mentally tougher" or "confident", or whatever made up narrative you want to tell yourself because you really understand "who knows how to win close games."
This thread is real, actual, interesting discussion. It was disguised the first page or so. Please carry on and kudos.
I don't agree with the blanket statement that "games that finish close are essentially random".
To use the poker analogy, you might be a much player than me overall, but if we're playing heads up, the hands that are close (and the bad beats) will be distributed randomly.
When we both push pre-flop, you might have me dominated half the time, but if you just look at the times when the board does one of us a huge favor (random trips or four to a flush), they will be distributed evenly.
In the basketball case, when one team has a real advantage, the game doesn't end within 2 points - they spend the last minute shooting FTs and push it to 8 or 10. The games that FINISH within one possession are much more subject to random variance.
Yea I understand all of that.
You can have two team of basically even strength that play to a 1 pt game where the ending involves a ton of random sequences. Ohio State versus Iowa State with that terrible call is an example. You can also have a close game where the randomness really doesn't have much of an impact on the final score. The Michigan vs. Kansas game is a great example of that imo. Those are two evenly matched teams and the final came down to one team (Burke mainly) making outstanding individual plays while the other team made bad plays (Elijah Johnson).
To make a blanket statement that games that end close are essentially random cheapens the amazing plays made by a guy like Trey Burke who really won the game for his team. He wasn't lucky there.
I know the whole idea of these type of stats and this way of thinking is to standardize certain situations and make everything uniform so you can study it but I'm not always a fan of it in the sport I played so maybe that is just a mental block for me.
I wasn't trying to say that their lack of success in close games was some type of conclusive sample size that proved they are a mentally weak team who cant win close games. That would be ridiculous. All I was saying was that players are more likely to tighten up and struggle down the stretch in a close game when they have lost all the close games they played in that season, and when teams have won more of their close games it is a lot easier to stay loose in those situations. I don't see how that goes against anything you are saying. There is obviously a lot of luck/chance/variance whatever you want to call it in every single close game. But by staying loose, or tightening up, players can "make their own luck" to an extent. Thats all Im trying to say
kenpom describes this better than I could. Two relevant links:
Bottom line: there is no observe predictive value in looking at a team's record in close games over a small sample size of 5 or so games. So if we want to argue about the nature of luck we can...but the bottom line is that it just doesn't matter that they've had a poor record in close games.
Good article about Florida in particular. Here's the money portion of it:
With this in mind, though, I found seven Florida wins against decent competition that had at least a moderate amount of tension at some point in the second half.
1. 3/16 vs. #61 Alabama (at Nashville) Trailed by 10 with 14 to go. Won by 10.
2. 3/2 vs. #61 Alabama Trailed by 8 with 12 minutes left. Won by 12.
3. 2/23 vs. #73 Arkansas Led by 4 at halftime. Won by 17.
4. 1/23 at #97 Georgia Trailed by 3 at halftime. Led by 14 with 10 to go. Won by 17.
5. 12/29 vs. #87 Air Force (at Miami) Led by 1 with 17 to go. Won by 17.
6. 11/23 vs. #138 UCF Led by 7 with 17:30 left. Won by 13.
7. 11/18 vs. #32 Middle Tennessee (at Tampa) Led by 4 with 16 minutes left. Won by 19.
If the Gators had just consistently struggled a little more down the stretch in these games, people could say they were 7-6 in close games and totally deserving of their high computer rankings. Or something like that.
To go back to my original Orioles example, the Os last year were 10-1 in the second half of the year in one run games. Obviously it took a lot of luck to achieve that record. But even if they got all the same lucky breaks that they did, I don't think they would have been able to put go 10-1 in one run games had they not been playing with confidence in those situations because of their success in those games earlier that season.
Am I making any sense?
Here's a good article on the subject, applied to football...scroll down to "Close Games"
Here are the four new-school football stats that attempt to predict the 2012 season.
First article provides solid data.
You are cherry-picking the data that fits your conclusion. You're selecting a 10 game stretch...in the second half of the year...because it fits what you want to believe.
Also, a one-run baseball game isn't nearly as "close" as a 2 or 3 point basketball game.
That was just an example. I don't think we're as far apart here as you seem to think. I have never once said variance doesn't play a large role in close games.
Do you disagree that experience winning close games makes it easier for a player to stay loose in a close game?
I think it might be semantics. I would never try to dismiss the existence of random other variables in basketball games. I know they exist.
I'm just saying that in a tie game with 2 minutes the winner of the game is almost impossible to determine. So yes in that sense the result is random. But the things that happen in that 2 minutes to determine the winner are "essentially random". Sometimes players step up and make key plays and sometimes the other teams players do it. But that isn't random to me.
Basically, I definitely agree that variance plays an enormous role in determining the outcome of close games. I do not agree that experience in past close games has no impact on the average player's mindset and performance in future close games. I don't see whats so ridiculous about such a statement
There is nothing ridiculous about that statement whatsoever. I would be willing to bet that almost all players that played the game at a high level would agree with you.
The reason someone would say it's ridiculous is that there is ton of data across a number of sports at a number of levels that we can look at to see how true this is. As far as I know, none of the many, many studies using that data have found the persistence of clutch performance to be anything more than minimal. So if it does exist, a) we can't find it and b) we can't use it to actually make predictions about the future.
Id guess that who wins a tie game with 2 min left isn't essentially random. There will be significant variance, but the better team will win more often if you play enough times. However, if you take out all the games where one team plays well enough in those 2 min to win by two possessions or more, and only count those that come down to the last possession, that subset of results are random.
Just like the poker example, you might win more often when we both push per flop because you push with better hands, but if you only look at the subset of hands where someone caught trips on the river to win, it will be random.
This post was edited by HoopheadVII 13 months ago
Now you're talking about individual player mindset (and performance). That's different from team performance, especially when that team is playing another team, officiated by humans. One player may make a difference, but there are 12 other people on the court that can affect the outcome.
Plus, if you exclude all the games where that one player makes enough of a difference for his team to win comfortably, the ones that are left look random.
I would like to see a similar breakdown as the kenpom article above, for teams that come back from a double digit deficit. The prevailing theory is that if they "get over the hump" like Michigan last night, they are destined to win in OT. I wonder if that's true.
Re: Trey Burke...
Home Court Live will track all of today's tournament action. Join in the conversation.
To answer Toogs's question, it would make more sense to analyze each individual case of that happening. What allowed for the team to come back from a double-digit deficit in the first place?
Was a player of theirs in foul trouble and on the bench while the other team's lead grew, and his return sparked the comeback? Conversely, was it foul trouble on the other end that changed the game?
Was it fatigue - did conditioning play a factor? Did an adjustment by the coach allow his guys to conserve energy or force the other team to expend more energy, in turn forcing the other team to play less skilled bench players OR keep their better players on the court while tired?
Did a home crowd help a struggling team find their rhythm? Or was this a veteran team that didn't get beaten down by opposing fans on the road?
Once we can examine the factors that led to the comeback, some will likely lead to a victory more than others. If the other team's star fouls out in regulation allowing for a comeback, we can easily see why that advantage would continue into the overtime period. If the coach was resting his star with 4 fouls, and he was able to play him the last 4 minutes of the game without drawing a foul, that would be a weaker indication that his team would win.
If you were looking to make wholesale determinations about who would win going into overtime, I think you'd want to ignore just about everything that happened in the game, actually, and look at talent. Looking at the specific Michigan-Kansas example and extrapolating, I'd want to know if a) Kansas developed a big lead because they are better/more talented than Michigan or did b) Michigan overcome that steep deficit because they're better/more talented than Kansas? If A, the Kansas choke/UM comeback is an unlikely event; odds of KU being outplayed in a five-minute interval again are slim. If B, I'd instead count on Michigan to do what they're supposed to do and simply outplay Kansas. It was the first 35 minutes that were an anamoly, not the last 5.
i dont even get why people watch sports anymore...just surf websites and youll know more than anyone
the internet has birthed a disgusting brethern of sport "fan"
i beg you all to come up to the princeton in ne philly on any saturday and introduce yourself to these things called televisions.....there will also be degenerate gamblers....alcoholics and dust heads.....but they all will actually watch the games until they are over...and they will do this without going on a single message board or website the entire day......its almost unfathomable
247Sports In partnership with CBS Sports