RebKell's Junkie Boards
Board Junkies Forums
 
Log in Register FAQ Memberlist Search RebKell's Junkie Boards Forum Index

Player evaluation

 
Post new topic   Reply to topic    RebKell's Junkie Boards Forum Index » WNBA
View previous topic :: View next topic  
Author Message
ClayK



Joined: 11 Oct 2005
Posts: 11105



Back to top
PostPosted: 04/23/17 11:02 am    ::: Player evaluation Reply Reply with quote

Perhaps the key to success at any level, player evaulation is always tricky. As noted by another poster, I rely heavily on statistics to determine a player's real or potential value, and not nearly as much on observation as others. Here are a couple of reasons why:

1) The top high school player in Northern California is widely considered to be 2019 wing Haley Jones of Archbishop Mitty. She is about 6-0, immensely skilled, a coach's daughter and plays for one of the top high school teams in the nation and one of the top club teams in the nation.

I have probably seen her play 10 times, at both levels and at USA Basketball. She has shown bursts of tremendous skill in those games, but not once I have seen her take over a game. She did not have good games at the Nike TOC, and did not dominate in California's postseason.

Is she an elite player? No question, and the statistics and other reports pretty much guarantee it. But if you were to ask me, off of my fairly extensive observation of her, I would not rate her as highly as others because of the sample of games I have seen her play.

I think this issue is a factor unless you basically see every game a player sees because even 10 games (or 15) is a pretty small sample size. And given the number of players who need to be seen, and the number of available games, I don't think anyone can see even half the games of elite players.

Now Jones is very, very good, and I don't accept my eye test as being definitive because of other factors -- but she's Exhibit A in why I don't trust what I see, unless I see a lot.

2) In high school, one year I had a starter who wasn't playing as well as I thought she could, and the girl who would sub for her seemed to come in and make us play better. The starter's stats were a little better, but I was about to make a change because of my perception of how we played with the backup in the game.

But that year, we had parents who were the best ever on statistics, and they were able to break down our team performance by which group of five were on the floor -- and those numbers showed that we were better with the starter. They didn't believe it, and neither did I, so they broke it down even further, making sure the bench player wasn't playing against weaker units from the other team. She wasn't.

So by the eye test, I would have made a change that would have hurt the team, as pretty much everyone agreed that the bench player was better. But the reality of what happened on the court trumped the eye test, and we stayed with the starter.

These kinds of evaluation issues crop up all the time, which is why I tend to rely more on statistics than seeing a player a few times, or even every time, than observation and perception. We all know about perception and reality, and when you add in our unavoidable personal biases about personality and style of play, I've found it's way too easy to make mistakes that a hard look at the numbers can avoid.



_________________
Oṃ Tāre Tuttāre Ture Svāhā
pilight



Joined: 23 Sep 2004
Posts: 66773
Location: Where the action is


Back to top
PostPosted: 04/23/17 11:46 am    ::: Reply Reply with quote

You can't just go on numbers. Observation has to be part of any serious evaluation. What you look for in an evaluation is different from what you look for as a fan or a journalist, however.



_________________
Let us not deceive ourselves. Our educational institutions have proven to be no bastions of democracy.
Randy



Joined: 08 Oct 2011
Posts: 10911



Back to top
PostPosted: 04/23/17 12:21 pm    ::: Reply Reply with quote

Numbers are data - pretty much objective. Eye test is subjective and bias can play role. For ex, its easy to trash a player's defense since the stats that measure it are incomplete and flawed.. Both have their place and limitations.


pilight



Joined: 23 Sep 2004
Posts: 66773
Location: Where the action is


Back to top
PostPosted: 04/23/17 12:27 pm    ::: Reply Reply with quote

Randy wrote:
Numbers are data - pretty much objective.


Numbers can fool you in many ways. When you start looking at college and below more advanced data often isn't available.



_________________
Let us not deceive ourselves. Our educational institutions have proven to be no bastions of democracy.
myrtle



Joined: 02 May 2008
Posts: 32326



Back to top
PostPosted: 04/23/17 1:05 pm    ::: Reply Reply with quote

One more bit of mud in the water. The level of competition makes a big difference in stats as well. Very hard to compute that, especially in high school. You can say that a player who dominates in the biggest school leagues in a state has the best chance at the next level. But the level of play also varies from state to state. And we have seen quite a few very good players who come out of smaller schools as well. Evaluating in comparison mode is very difficult.

Going from college to the pros is a little easier because we have a good idea about the strength of the competition. And it is harder for a mid-major to make it in the pros. Exceptions like EDD of course exist, but she's a pretty unique case.



_________________
For there is always light,
if only we’re brave enough to see it.
If only we’re brave enough to be it.
- Amanda Gorman
ClayK



Joined: 11 Oct 2005
Posts: 11105



Back to top
PostPosted: 04/23/17 3:07 pm    ::: Reply Reply with quote

High school stats are pretty much worthless, especially since many are kept by parents. And the level of competition is important, of course.

And a combination of evaluations are obviously the best, since first, there's a bar of athleticism and size that must be exceeded to play at the next level. It's possible to excel at one level but simply not have the physical tools to do as well up one rung of the ladder.

But shooting percentage, especially from three-point distance, is a pretty solid predictor, as is A/TO. FT% also gives an idea of the player's raw shooting ability.

I also look at rebounds per minute played, as that's a pretty good indication of rebounding ability, though of course different positions mean different results mean different things.

Defensive stats are not good, and not even present in college and high school, so there, watching games is key. But I think you have to watch very, very carefully, as even a decision like playing zone vs. man may reflect on a particular player's defensive ability -- or not. If a team is in man, then observing who the coach matches up the player in question with is information in and of itself, on top of which comes the on-ball defense.

And on-ball defense is different than help defense, or rim protection, or even trapping ability.

But perimeter players at the WNBA level, barring the occasional Danielle Robinson, must be able to score from the perimeter to be effective, and so college three-point shooting is a valid evaluation tool. And ballhandling at every position is critical, so A/TO says a lot (again, differing by position).

What we're looking for is predictive information, and I'm not convinced the eye test is more accurate than certain stats. For WNBA evaluation, perimeter players almost always need to make at least 35% of their college threes to be successful, and players who are expected to rebound must get at least one every four minutes.

Tiffany Mitchell, for example, looks like a WNBA player, but I'm yet to be convinced she is. Brittany Boyd does too, but still has a ways to go, it seems.



_________________
Oṃ Tāre Tuttāre Ture Svāhā
pilight



Joined: 23 Sep 2004
Posts: 66773
Location: Where the action is


Back to top
PostPosted: 04/23/17 3:40 pm    ::: Reply Reply with quote

ClayK wrote:
But shooting percentage, especially from three-point distance, is a pretty solid predictor, as is A/TO. FT% also gives an idea of the player's raw shooting ability.


See, now you've fooled yourself into thinking Curtyce Knox and Breyana Mason are better prospects than Kaela Davis. While I think Davis is overrated, she's more of a prospect than her stats indicate.



_________________
Let us not deceive ourselves. Our educational institutions have proven to be no bastions of democracy.
sigur3



Joined: 18 Jun 2013
Posts: 6191
Location: Chicago-ish


Back to top
PostPosted: 04/23/17 4:39 pm    ::: Reply Reply with quote

You need context for your numbers. Especially when talking about college players.


Angus24



Joined: 13 Nov 2007
Posts: 686
Location: South Dakota


Back to top
PostPosted: 04/23/17 4:46 pm    ::: Reply Reply with quote

Not much said about physical characteristics in these discussions. Size and speed can be crucial sometimes helping and other times hurting in the evaluation process. I think Becky Hammon was a classic example of that. She had tremendous stats coming out of college but went undrafted because of her size and speed. That draft was unusual of course because of all the ABL players entering the draft. If you believe Richie Adubato the thing that attracted him was her shooting skill which evidently trumped everything else in his mind. I assume he saw ways of making use of that exceptional skill even if she was small and slow compared to other point guards. He certainly proved to be right.


ClayK



Joined: 11 Oct 2005
Posts: 11105



Back to top
PostPosted: 04/23/17 5:55 pm    ::: Reply Reply with quote

As I began my post
Quote:
there's a bar of athleticism and size that must be exceeded to play at the next level. It's possible to excel at one level but simply not have the physical tools to do as well up one rung of the ladder.


That's step one. The stats don't matter if you don't pass that test.

If you do, then the stats can help. My issues with Kaela Davis are consistency and defense, neither of which can be answered from her college play, I don't think.



_________________
Oṃ Tāre Tuttāre Ture Svāhā
Richyyy



Joined: 17 Nov 2005
Posts: 24326
Location: London


Back to top
PostPosted: 04/23/17 6:10 pm    ::: Reply Reply with quote

ClayK wrote:
My issues with Kaela Davis are consistency and defense, neither of which can be answered from her college play, I don't think.

So no issues with the godawful shooting percentages?

On the more general subject, anyone with half a brain will tell you that you need to combine the numbers with watching the games to get a realistic view. Only the geekiest of basement-dweller will claim the numbers can tell you everything, and only the dumbest 'old school' guy will dismiss the numbers entirely in favour of what he claims to see. They need to inform each other.



_________________
Independent WNBA coverage: http://www.wnbalien.com/
sigur3



Joined: 18 Jun 2013
Posts: 6191
Location: Chicago-ish


Back to top
PostPosted: 04/23/17 7:01 pm    ::: Reply Reply with quote

Richyyy wrote:

On the more general subject, anyone with half a brain will tell you that you need to combine the numbers with watching the games to get a realistic view. Only the geekiest of basement-dweller will claim the numbers can tell you everything, and only the dumbest 'old school' guy will dismiss the numbers entirely in favour of what he claims to see. They need to inform each other.


Yes, but that also requires common sense.


RavenDog



Joined: 19 Feb 2007
Posts: 6863
Location: Home


Back to top
PostPosted: 04/24/17 8:43 am    ::: Reply Reply with quote

pilight wrote:
ClayK wrote:
But shooting percentage, especially from three-point distance, is a pretty solid predictor, as is A/TO. FT% also gives an idea of the player's raw shooting ability.


See, now you've fooled yourself into thinking Curtyce Knox and Breyana Mason are better prospects than Kaela Davis. While I think Davis is overrated, she's more of a prospect than her stats indicate.


Razz A change of heart here with Kaela perhaps?


ClayK



Joined: 11 Oct 2005
Posts: 11105



Back to top
PostPosted: 04/24/17 9:31 am    ::: Reply Reply with quote

Kaela Davis, at her best, is a well above average WNBA offensive player, which is worth quite a bit. But ...

1) She's not well above average every time out.

2) Her ability to defend at the three in the WNBA is a mystery.

3) And though it was several years ago, I have reservations about any player who quit on her high school team just before postseason began. That kind of action calls into question the desire to fit into a team format. Now maybe she's grown up since then and looks back on that as a very big mistake, in which case it's not an issue. But still, how many kids do that?



_________________
Oṃ Tāre Tuttāre Ture Svāhā
Jet Jaguar



Joined: 11 Feb 2014
Posts: 1111



Back to top
PostPosted: 04/24/17 10:05 am    ::: Reply Reply with quote

I think you should have started the sub on your team and rely on your instincts. If it didn't work out, switch back. If the player looked better they maybe were, just needed a different more prominent role. Stats are so deceiving. Can be affected by team makeup, role, level of competition, chemistry, etc, etc. Skill is unaffected by anything. You just see it. Sometimes kids just need a chance to prove themselves. That 2nd scenario didn't prove anything except that you didn't try the change, based on stats rather than what you saw. So you don't really know if what you did was the right decision or not. You just assumed it was based on some stats.

Obviously you need to take into affect both stats and eye test, but to rely solely on stats like some seem to do, produces some pretty poor player evaluations.



_________________
Oderint dum metuant - Let them hate, so long as they fear
ClayK



Joined: 11 Oct 2005
Posts: 11105



Back to top
PostPosted: 04/24/17 10:34 am    ::: Reply Reply with quote

Jet Jaguar wrote:
I think you should have started the sub on your team and rely on your instincts. If it didn't work out, switch back. If the player looked better they maybe were, just needed a different more prominent role. Stats are so deceiving. Can be affected by team makeup, role, level of competition, chemistry, etc, etc. Skill is unaffected by anything. You just see it. Sometimes kids just need a chance to prove themselves. That 2nd scenario didn't prove anything except that you didn't try the change, based on stats rather than what you saw. So you don't really know if what you did was the right decision or not. You just assumed it was based on some stats.

Obviously you need to take into affect both stats and eye test, but to rely solely on stats like some seem to do, produces some pretty poor player evaluations.


I guess the point here is objective vs. subjective, and in the case of the starter (failed eye test) vs. sub (failed the stats test), it's not as if the sub didn't get a chance to play, or didn't get a chance to play with the other starters. She had plenty of opportunities to prove herself, and did, getting more and more minutes as the season went on.

Over the past 20 years, the trend in sports has been to look more closely at the numbers than go with the feel (sacrifice bunts in baseball, midrange jumpers in basketball, etc.), and Darwinian results land on the side of the numbers.

But obviously there is a place for subjectivity and feel, but in general, if you value wins and losses, objectivity gets better results.



_________________
Oṃ Tāre Tuttāre Ture Svāhā
PUmatty



Joined: 10 Nov 2004
Posts: 16346
Location: Chicago


Back to top
PostPosted: 04/24/17 1:03 pm    ::: Reply Reply with quote

The problem with statistics, and any type of prediction, is that you have to have enough data to understand what is actually predictive of success. Basically, you have to know that the statistic - advanced or not - that you are looking at is actually predictive of wins. And to do that, you have to have shit-ton of data (that's the technical term). This is why MLB is such a natural place to have these kind of data - you have so many games a year and so many years of data that you can estimate much more complex models than other sports.


ArtBest23



Joined: 02 Jul 2013
Posts: 14550



Back to top
PostPosted: 04/24/17 2:44 pm    ::: Reply Reply with quote

ClayK wrote:

Over the past 20 years, the trend in sports has been to look more closely at the numbers than go with the feel (sacrifice bunts in baseball, midrange jumpers in basketball, etc.), and Darwinian results land on the side of the numbers.

But obviously there is a place for subjectivity and feel, but in general, if you value wins and losses, objectivity gets better results.


Well, there are limits. For example, since the NFL moved back the distance for extra points, the statistics show that the points-per-attempt yield is actually slightly higher for 2 point conversions than for kicks. But you don't see the entire league abandoning the single extra point attempt.


Jet Jaguar



Joined: 11 Feb 2014
Posts: 1111



Back to top
PostPosted: 04/24/17 6:08 pm    ::: Reply Reply with quote

ClayK wrote:
Jet Jaguar wrote:
I think you should have started the sub on your team and rely on your instincts. If it didn't work out, switch back. If the player looked better they maybe were, just needed a different more prominent role. Stats are so deceiving. Can be affected by team makeup, role, level of competition, chemistry, etc, etc. Skill is unaffected by anything. You just see it. Sometimes kids just need a chance to prove themselves. That 2nd scenario didn't prove anything except that you didn't try the change, based on stats rather than what you saw. So you don't really know if what you did was the right decision or not. You just assumed it was based on some stats.

Obviously you need to take into affect both stats and eye test, but to rely solely on stats like some seem to do, produces some pretty poor player evaluations.


I guess the point here is objective vs. subjective, and in the case of the starter (failed eye test) vs. sub (failed the stats test), it's not as if the sub didn't get a chance to play, or didn't get a chance to play with the other starters. She had plenty of opportunities to prove herself, and did, getting more and more minutes as the season went on.

Over the past 20 years, the trend in sports has been to look more closely at the numbers than go with the feel (sacrifice bunts in baseball, midrange jumpers in basketball, etc.), and Darwinian results land on the side of the numbers.

But obviously there is a place for subjectivity and feel, but in general, if you value wins and losses, objectivity gets better results.

One other factor I forgot to mention about players productivity is confidence. And confidence is often fed off of a coaching believing in that player. I've heard someone once say "You (a player) are only as good as your coach thinks you are" (it was Jim Foster in a recent interview). So switching a player from backup to starter who looks better in the eye test may be all a player needs to catapult them to the next level. Sometimes that vote of confidence is all a player needs.



_________________
Oderint dum metuant - Let them hate, so long as they fear
myrtle



Joined: 02 May 2008
Posts: 32326



Back to top
PostPosted: 04/24/17 6:23 pm    ::: Reply Reply with quote

Jet Jaguar wrote:
ClayK wrote:
Jet Jaguar wrote:
I think you should have started the sub on your team and rely on your instincts. If it didn't work out, switch back. If the player looked better they maybe were, just needed a different more prominent role. Stats are so deceiving. Can be affected by team makeup, role, level of competition, chemistry, etc, etc. Skill is unaffected by anything. You just see it. Sometimes kids just need a chance to prove themselves. That 2nd scenario didn't prove anything except that you didn't try the change, based on stats rather than what you saw. So you don't really know if what you did was the right decision or not. You just assumed it was based on some stats.

Obviously you need to take into affect both stats and eye test, but to rely solely on stats like some seem to do, produces some pretty poor player evaluations.


I guess the point here is objective vs. subjective, and in the case of the starter (failed eye test) vs. sub (failed the stats test), it's not as if the sub didn't get a chance to play, or didn't get a chance to play with the other starters. She had plenty of opportunities to prove herself, and did, getting more and more minutes as the season went on.

Over the past 20 years, the trend in sports has been to look more closely at the numbers than go with the feel (sacrifice bunts in baseball, midrange jumpers in basketball, etc.), and Darwinian results land on the side of the numbers.

But obviously there is a place for subjectivity and feel, but in general, if you value wins and losses, objectivity gets better results.

One other factor I forgot to mention about players productivity is confidence. And confidence is often fed off of a coaching believing in that player. I've heard someone once say "You (a player) are only as good as your coach thinks you are" (it was Jim Foster in a recent interview). So switching a player from backup to starter who looks better in the eye test may be all a player needs to catapult them to the next level. Sometimes that vote of confidence is all a player needs.


or vice versa. The starter getting benched may lose all confidence and start playing worse even if you then put her back in the starting lineup. Or that starter being benched might get inspired and start playing better. Coaches have to be psychologist as well as coach (which is one of the reasons Geno is so amazing) - know when to push 'em, know when to pat 'em on the back, know when to bench 'em, etc...



_________________
For there is always light,
if only we’re brave enough to see it.
If only we’re brave enough to be it.
- Amanda Gorman
Display posts from previous:   
Post new topic   Reply to topic    RebKell's Junkie Boards Forum Index » WNBA All times are GMT - 5 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB 2.0.17 © 2001- 2004 phpBB Group
phpBB Template by Vjacheslav Trushkin