If the NCAA was a series of 7 game tournaments, all UCLA would’ve done by winning the first game is improve their odds of getting to 6 games rather than losing in 5 against Michigan and Bama. They’d be swept by Gonzaga.
4 Likes
I don’t even understand the point of this to be honest - they won the game. It’s a single elimination tournament, and single game contests aren’t predictive - bad teams in every league win games. UCLA won 5 games in a row and beat some very good teams (yes by slim margins) - we can talk about FT %, but part of being a good team is playing well - like making free throws. This desire to not give them credit doesn’t make sense to me. The game isn’t played in a calculator - taking better shots, etc., matters but it doesn’t overturn the result…and the result is what the world is going to remember.
3 Likes
I think it all just stemmed from whoever said that Michigan’s win probability when they were up 7 ten minutes into the game wasn’t accurate. UCLA had <25% of beating each of Michigan, Alabama, and Gonzaga. They beat Michigan and Alabama, but that doesn’t mean their pre-game win probabilities were wrong.
2 Likes
It’s “accurate” but that doesn’t mean it’s meaningful. 7 point first half leads aren’t a bullet proof vest.
2 Likes
Nor are 14 point leads, ask Michigan State! 
Hence, the 25%…
4 Likes
Don’t think anyone said it was bullet proof. That’s why it wasn’t 100%
3 Likes
This talk of “luck” I just think is silly. Last night Skip Bayless (I know, I know) said Jalen Suggs’ shot was “lucky”. In the sense the odds are against that shot going in, I guess it was. But describing the results of highly trained basketball players attempting to make shots as “luck” seems to miss the point. 45% shooters don’t churn out 45% shooting nights like a metronome.
Right but it’s heavily weighted by a pre-game resume that was built with the presence of Isaiah Livers.
2 Likes
I made a comparison with making a hole in one in a conversation today. There is certainly a huge element of luck in making a hole in one but generally those with great skill tend to be a lot “luckier” than those lacking in skill. You hit a shot to a hole 180, 190, 200 yards away and somehow that ball falls into the hole. Good players are going to have the skill to get the ball near the hole, on line, correct distance, etc, but there is certainly some luck in having the ball fall in the hole.
With Suggs last night he had the ability to dribble it quickly up the floor, to keep his head up, to launch a shot from what, 40 feet, and have that shot be online and at approximately the right distance. I’m pretty sure he didn’t mean to bank it so it certainly wasn’t the perfect distance, but rather it was about what, eight, ten, twelve inches too long? In fact if that ball had been just a couple of inches longer or shorter than the distance it traveled, if it had been a few inches off line either way, it likely wouldn’t have gone in. Tremendous skill got it near the basket, but a little luck got it to go in. Doesn’t mean it wasn’t a GREAT shot!
3 Likes
Right - there is some element of luck in al most anything that happens in the sports arena. Sometimes very little, sometimes a lot. But “unlikely” and “lucky” aren’t the same.
1 Like
If you take a bunch of low percentage shots and make them consistently, they cease being low percentage, bad shots.
People are right to question that 88% chance of winning in the first 6 mins of the game statistic. None of the predictive stats used by ESPN for games or by Newspapers for elections are very sophisticated not only because they fail to account for a bunch of different factors that aren’t quantifiable but because that ‘reams’ of data simply doesn’t exist.
What they don’t tell you is the margin of error for that prediction is massive (like, 20%). You can say there’s 100s of years worth of historical data but the sample size is much much smaller when you take all the number 1 seeds that faced 11 seeds, all the times the scoreline was exactly that or within that margin at that specific period of the game. That sample size is probably pretty small, nevermind the fact that you have to standardize what a 1 or 11 seed meant across all the years of data from which you got your samples from.
Never mind the historical anomaly this year ways, making blanket statements like UCLA was purely lucky or that we are an overwhelmingly better team is just not right.
Edit: Also, there are so many hypotheticals given for the argument for why UCLA was so lucky. If Alabama was shooting their normal amount of FT, If Michigan did this, if Zags did that…way too many ifs
2 Likes
No, they don’t. Luck can persist for a length of time; it doesn’t mean that it’s not luck. You can roll eight consecutive pairs of 3s, but that doesn’t mean that betting on 6 the hard way has increased its EV.
That’s… not how those calculations are done.
Win probability is a straightforward stat to calculate, and it doesn’t depend upon any of those parameters. The inputs are the current margin, an estimated amount of time / number of possessions remaining, and the distribution of points per possession for each team.
4 Likes
That was me. But this is really tangential to my larger point, which is that the ESPN author acted like anything short of the Final Four was a debacle for Michigan. It’s not clear to me if the author even knew we played the tournament missing one of our best players.
Personally, I thought overall we responded well to Livers’ absence and Juwan showed his chops as a tourney coach. The idea that we should be totally down right now (like the author suggested) doesn’t make sense to me.
6 Likes
Yes, but there certainly IS an element of luck. Let’s not argue semantics. Suggs is a VERY skilled shooter who made a wonderful shot to win at the buzzer. He timed it perfectly. His brain calculated it almost perfectly, on the run, heat of the battle, head up, eyeing the bucket, and he let it fly, just a bit long, but just the right amount too long, and it careened off the backboard from the perfect spot on that backboard, and it went in the basket. Game. Wow, so much skill and talent to get it so close, close enough to hit the backboard at the perfect angle to go in. Tremendous skill and just a little luck.
Now, if you told me he called “backboard” I’d say there was no luck involved. The fact that there was just a little luck does not detract from the skill Jalen Suggs was able to call upon in that moment for that game winning shot. I’m glad we agree that while there was great skill involved, by an amazing player, there was “some element of luck.”
I will say this, though, too, when you have a great player like Jalen Suggs there is much less luck involved than if, for instance, I was shooting that shot! If I had been shooting in that situation, and I wouldn’t ave been, it would have been almost ALL luck! 
1 Like
So in the context of UCLA it wasn’t so much luck as just extreme outlier level of shooting. They were making the shots which is skill, it’s just based on their averages it was unsustainable in many way. But all the credit goes to guys like Juzang and Campbell for just totally balling out.
On the Gonzaga yesterday side it was “unlucky” that UCLA hit so many tough shots at such a high percentage.
2 Likes
Yes basically everything that happens on a court is on a sliding scale of “luck” - we could call missing a dunk “unlucky”. What I’m saying is that this stuff happens, and it’s why basketball isn’t just performed by putting KenPom profiles in a computer. To borrow a cliche - “That’s why they play the game.”
1 Like
The probabilities are built on historical data. A team with x point lead with y left on the clock has won this percentage of times. Efficiency data (that’s been accurate in picking games before they start) is then factored in to make the probabilities more accurate.
You don’t need to account for a lot of factors. Just the above is enough to give an accurate percentage.
Some of you are getting too swayed by the final outcome. 78% chance of victory means 22% chance of defeat.
2 Likes
Even KenPom isn’t particularly luck-invariant, FWIW. Far better (in terms of being predictive and also in terms of judging overall team quality) would be a system that computed the expected value of each shot and used that instead of its actual result.
That’s this site (I don’t know how accurate it’s pregame predictions are against the spread.)
3 Likes