Not an UnderStatenment, Cats take a turnover for the worse; WVU 2 ADV

[michigancat]

This is going back to an early conversation, but ever since I've seen kenpom's stuff I've have sys' mentality. It sucks that we lost at UT because that is an "L" on our record that we can't get back. But it's good to know that based out what I saw and what the stats show, we probably win that game 5 out of 10 times. I think sys shares this perspective, and that's why he didn't see that loss as a big deal. It gives me confidence in our team going forward.

The thing that stats can not take into account is how a bunch of 18-19 year olds react to 3 losses in the past 4 games. They may turn it off for the season and start preparing for the next season or just start trying to make it through practices for the rest of the year instead of getting better like they had been earlier this season. I don't think that's the case. My point is that even if michigancat and sys don't believe it or just don't want to believe it, emotion and things like "clutchness" are factors to some extent.

I would say they are only factors if they can be meaningfully measured.   If "clutchness" doesn't result in an actual observable (such as more/better scoring  [how ever you want to measure such a thing]) with a numerical value, then I personally doubt its an effective factor.

I think there is a difference between being clutch and not being poised. I don't think clutch is measurable or even exists. I do think you can look to things like TOr in late and close situations to gauge whether or not a team has poise.

You could, but you'd need a far larger sample than what we're working with to discern anything out of the ordinary.

also, LOL at differentiating between "clutchness" and "poise". What on earth are you talking about, MiR.

[CHONGS]

The distinction isn't numbers vs. words.  Numbers can be expressed as words.  It's just a matter of rigor, which some people screw up on even when they try to use stats.

I would say it's measurable vs unmeasurable. 

During a basketball game there are a large number of "objective" observables, if you claim "poise" is one of them, then I would expect a well-defined metric in which to measure this.  Anyone in the crowd should be able to measure this, and all measurements should be essentially the same (after accounting for errors in measurement and systematic error).  For example, the number of points scored in a game is something I would call an objective observable.  Even if someone in the crowd is taking score and they miss a score (a measurement error) or the official makes a mistake (systematic error) this can eventually be accounted for because before the game started it was agreed upon what constitutes a score (by the rule book).  So can other things like rebounds, FTR, etc.   Something like jersey color is an objective observable as well so are number of vowels in a player's name.  These last two can be measured, but the next step would be to show if/how they affect the other observables (especially observables that are generally considered important like wins, points, etc..).   

Ok a lot of stuff and I will generally stop now before it gets too boring.  If anyone wants to keep on this topic, however, I do like talking about it.

 

[sys]

The distinction isn't numbers vs. words.  Numbers can be expressed as words.  It's just a matter of rigor, which some people screw up on even when they try to use stats.

a number expressed as a word is still a number.

[sys]

i tried to think of something that could be measured that wasn't numerical for about five minutes before my original post, btw.  i couldn't think of anything.  if someone else can, i'd like to know what it is.

[chum1]

I think the standard concept of measurable includes numerical.

[chum1]

How much different, though, is saying "I want this team to play better than their average for the last few minutes of a game," from "I want this team to win the Big 12"?  I realize this isn't what people were saying, but maybe it's in the ballpark, and perhaps it's not all that crazy.  It seems much less crazy than saying, "I want this team to win all of its games by a margin of at least 60 points."

[michigancat]

How much different, though, is saying "I want this team to play better than their average for the last few minutes of a game," from "I want this team to win the Big 12"?

"better play over the last few minutes of a game" will have a lot more variance/luck involved because it's a much smaller sample size. Winning the Big 12 is also much easier to define/measure.

[gatoveintisiet]

Stats-what do I do with them? Are they predictive of future events? can they give me a statistically sound advantage over a bookie?
Do they just describe what happened in the game that I just watched?  JMHO but if no significant future advantage can be gained
I'm gonna sit over here in camp Tubesock where I look at the after the fact stats and say to myself "yep thats pretty much what
I just saw on tv, but wow I didn't realize they went to the line THAT much more than we did" basically woopty-effing-doo.

[kso_FAN]

Predictive: if we allow 1.10 ppp and play games at 70+ possessions we will not win another road game this year. We were fortunate against ISU and WVU that we can even talk about turnovers being a factor as poorly as we controlled pace and played defense.

[michigancat]

How much different, though, is saying "I want this team to play better than their average for the last few minutes of a game," from "I want this team to win the Big 12"?

"better play over the last few minutes of a game" will have a lot more variance/luck involved because it's a much smaller sample size. Winning the Big 12 is also much easier to define/measure.

To expand, I don't think we should be concerned with late-close-road game mental toughness, we should be concerned with our inability to develop a meaningful lead in these games (or any lead at all in the case of WVU and ISU). If I were a coach, I wouldn't spend any time working on trying to correct the 37 late-game possessions we've talked about or develop mental toughness,- I'd spend time looking for trends in the other 150 or so possessions in those games and seeing what I can correct there. I think this is the heart of what this discussion is about.

[gatoveintisiet]

Predictive: if we allow 1.10 ppp and play games at 70+ possessions we will not win another road game this year. We were fortunate against ISU and WVU that we can even talk about turnovers being a factor as poorly as we controlled pace and played defense.

will we or will we not?  I don't think anyone has the answer to that right? Fan I know you love statistics and don't think it's
bad or anything, but I am curious how they can be useful to me as a fan.

[kso_FAN]

How much different, though, is saying "I want this team to play better than their average for the last few minutes of a game," from "I want this team to win the Big 12"?

"better play over the last few minutes of a game" will have a lot more variance/luck involved because it's a much smaller sample size. Winning the Big 12 is also much easier to define/measure.

To expand, I don't think we should be concerned with late-close-road game mental toughness, we should be concerned with our inability to develop a meaningful lead in these games (or any lead at all in the case of WVU and ISU). If I were a coach, I wouldn't spend any time working on trying to correct the 37 late-game possessions we've talked about or develop mental toughness,- I'd spend time looking for trends in the other 150 or so possessions in those games and seeing what I can correct there. I think this is the heart of what this discussion is about.

This.

"It is Groundhog Day for us. We keep repeating the same mistakes on the road. We do not play very well in the first half, but are able to battle in the second half to give ourselves a chance."

[MakeItRain]

How much different, though, is saying "I want this team to play better than their average for the last few minutes of a game," from "I want this team to win the Big 12"?

"better play over the last few minutes of a game" will have a lot more variance/luck involved because it's a much smaller sample size. Winning the Big 12 is also much easier to define/measure.

To expand, I don't think we should be concerned with late-close-road game mental toughness, we should be concerned with our inability to develop a meaningful lead in these games (or any lead at all in the case of WVU and ISU). If I were a coach, I wouldn't spend any time working on trying to correct the 37 late-game possessions we've talked about or develop mental toughness,- I'd spend time looking for trends in the other 150 or so possessions in those games and seeing what I can correct there. I think this is the heart of what this discussion is about.

You don't really think I was advocating that the coaches should be working on late game situations do you? I was simply explaining why IMO why we were/are garbage in such situations. Obviously all of these abstract intangibles we've talked about, if they exist, can't be fixed in practice. As a coach it would be more prudent to work on things that can be fixed like shooting, rebounding, ball handling, and rotational defensive covers. Although in February I really doubt that coaching can have measurable impact on any of those things either.

[kso_FAN]

Predictive: if we allow 1.10 ppp and play games at 70+ possessions we will not win another road game this year. We were fortunate against ISU and WVU that we can even talk about turnovers being a factor as poorly as we controlled pace and played defense.

will we or will we not?  I don't think anyone has the answer to that right? Fan I know you love statistics and don't think it's
bad or anything, but I am curious how they can be useful to me as a fan.

Many times my perception of what happens in a game is not what really happened and the numbers back that up. Now I generally track the stats as I watch, so that happens less. For me the numbers add to and help illuminate what I see on the court. That's just how I watch basketball now.

[michigancat]

Predictive: if we allow 1.10 ppp and play games at 70+ possessions we will not win another road game this year. We were fortunate against ISU and WVU that we can even talk about turnovers being a factor as poorly as we controlled pace and played defense.

will we or will we not?  I don't think anyone has the answer to that right? Fan I know you love statistics and don't think it's
bad or anything, but I am curious how they can be useful to me as a fan.

You can't say with any certainty, but you could say there is a decent chance of this happening at Baylor or OSU, based on our results and their results. (Actually there's a good chance we won't hit that pace against Baylor, but I don't think a fast pace hurts this team at all).

[michigancat]

You don't really think I was advocating that the coaches should be working on late game situations do you? I was simply explaining why IMO why we were/are garbage in such situations.

I honestly have no idea what the eff have you been advocating. I think you just like to argue with me.

[cDubya]

Man, I can tell it's offseason...

[sys]

Stats-what do I do with them? Are they predictive of future events? can they give me a statistically sound advantage over a bookie?

if you don't use stats, the book has a significant advantage over you.  actually the other side of the bet does, not the book.

[Fake Sugar Dick (WARNING, NOT THE REAL SUGAR DICK!)]

The only person arguing against the stats in this thread is Michigancat

[chum1]

How much different, though, is saying "I want this team to play better than their average for the last few minutes of a game," from "I want this team to win the Big 12"?

"better play over the last few minutes of a game" will have a lot more variance/luck involved because it's a much smaller sample size. Winning the Big 12 is also much easier to define/measure.

To expand, I don't think we should be concerned with late-close-road game mental toughness, we should be concerned with our inability to develop a meaningful lead in these games (or any lead at all in the case of WVU and ISU). If I were a coach, I wouldn't spend any time working on trying to correct the 37 late-game possessions we've talked about or develop mental toughness,- I'd spend time looking for trends in the other 150 or so possessions in those games and seeing what I can correct there. I think this is the heart of what this discussion is about.

I think it's no less legitimate for people to have an aesthetic preference for what that they perceive to be less sloppy play at the end of a game.  And what I was trying to say above is that I don't think this preference is relatively unrealistic.

[gatoveintisiet]

Stats-what do I do with them? Are they predictive of future events? can they give me a statistically sound advantage over a bookie?

if you don't use stats, the book has a significant advantage over you.  actually the other side of the bet does, not the book.

Really? Explain please

[michigancat]

I think it's no less legitimate for people to have an aesthetic preference for what that they perceive to be less sloppy play at the end of a game. 

Less legitimate than what?

And what I was trying to say above is that I don't think this preference is relatively unrealistic.

I think it's unrealistic to expect much control over the preference of less late-game sloppiness (without seeing significant changes overall). Yeah, I'd prefer we perform much better at the end of close games, but I can accept there's not much that can be done about it, and therefore not a cause for concern.

[sys]

Stats-what do I do with them? Are they predictive of future events? can they give me a statistically sound advantage over a bookie?

if you don't use stats, the book has a significant advantage over you.  actually the other side of the bet does, not the book.

Really? Explain please

a small number of bettors may have statistical models that are better than the statistical models used to set the lines.  tons of betters have access to (and use) models that are only a little bit worse.  to the extent that sophisticated bettors can overperform, it comes from bettors who underperform.  that group will include both those general bettors that use inferior statistical models and those that don't make use of any statistical information.

[pissclams]

Man, I can tell it's offseason...

no.  these guys love this crap.

[kso_FAN]

Man, I can tell it's offseason...

no.  these guys love this crap.

Yes!