In the world of analytics, there is no choice to be made. It is either a 3 or a lay-up. Mid-range game is dead, foul shots are devalued, weakside defense is exposed and all the traditional pillars of the game have been shaken to its core.

Scouting reports are no longer visual. It doesn't matter what you see unless a number backs it up. No decisions are made without consulting the stats.

Game plans are less about the game and more about the analytical matrix.

And, the higher up you go, the more analytically driven you are.

At lower levels, coaches have less of an understanding of the meaning and implications of what they think they are supposed to do.

In truth, the analytics do not paint the same picture at all levels and that is why I have a concern about younger players really learning the game so they can make sound decisions, as they get older.

I also have a concern about young coaches growing into their jobs with a firm understanding of what the game is about.

Please understand that I am not making (or at least trying not to make) value judgments of anyone's coaching. Coaches can coach however they like.

We all have different experiences that lead us to the decisions we make. Ball goes in, great decision. Ball bounces off the front rim, go find another job. The situation is the same, the result is different. Everything works, but everything doesn't work. You have to be comfortable with what you do.

Rather, I am examining the interpretation of the analytics that we coach by. Actually, I am examining the application of those analytics in one particular situation.

In this day and age, two of the most important statistics are "Points Per Possession" (the value of each shot you take) and "Effective Shooting Percentage" (shooting percentage that figures in 2 and 3 point shots). Below are the formulas.

Points Per Possession

Points / (FGA + (.44 x FTA) + TO)

If you want to learn more about Points Per Possession click this link
https://www.nba.com/resources/static/team/v2/thunder/statlab-en-october-1819-update.pdf

Effective Shooting Percentage

(FG + 0.5 * 3P) / FGA

When it comes to statistic, according to the tons of stats courses I took in college, there is a "Level of Significance" that needs to reached in order for a statistic to become valid and reliable.

In basketball, that Level of Significance is: 100 possessions. When used, the figures are always extended to 100 possessions.

Therein lies the issue, especially at lower levels. Unless you can track at least 100 possessions, the statistics are not telling you what you think they are. The more possessions you can track, the more reliable your measurement becomes.

This is a really long way around to the issue I really would like to address.

Is This the Last Shot You Want?

I go to a lot of games, games of all levels. I also try to go to games where I know the coaches so I can talk a bit about the game. I recently went to a junior college game. We have really good JC basketball here. Most of the teams are Div. I and a lot of players move on to play at 4-year schools.

There is a school about 15 minutes from me and I see them often. They are pretty good. Their coach prefers to press, trap and shoot a lot of shots. Like most things, sometimes it works out pretty well, other times not.

When they don't play well, it is usually for one of two reasons, they give up too many layups or they shoot themselves out of games with horrible shot selection.

In this particular game, my local team was playing the #1 team in the region. It was a great game, up and back, matching shot for shot, great intensity and competitiveness.

When it was all said and done, the local team was down 2 with 7 seconds to go and a throw in from the sideline. They inbounded the ball, peeled a player off a weakside stagger screen. The player broke free, ran 2 steps above the 3-point line, caught it and shot an air ball. Buzzer rang, game over, they lose by 2. That happens, that's the game.

I have seen them play a lot. The shooter was 6-6, strong and very athletic. He can shoot it ok but really has great success around the rim. When he puts the ball on the floor and gets into the lane, he's really a handful. He also likes to shoot 3's.

After the game, I went down to talk to the coach, whom I have known since he was a high school player. I asked him, "Did you get the shot you wanted?" He said, "Yes." I'm not making a judgment or a criticism. They were down 2 against a very good team, the longer the game went the more the advantage goes to the better team. Their best player had fouled out so they were trying to win in regulation rather than go into overtime. I might have made the same decision.

I then asked him, "Did you get the shooter you wanted." He said, "Yes." He explained to me that he had made a couple during the game so they thought he could make the shot. We think he's a pretty good shooter and he would be a good choice."

I had seen them play at least 8 times this year. He was their 2nd leading scorer. I remember the shooter as a sporadic, volume shooter from 3 and a very effective player on the drive and off the glass. When I looked at his stats they revealed he was a 43% shooter from the field and a 23% shooter from 3, with a significant number of attempts. Most of his makes come off kickbacks from inside the lane, mostly from the top.

Where do analytics fit in here? The first thing that jumped out at me was 43% from the floor with 23% from 3. That's significant. That says to me that he makes a lot of shots when he doesn't shoot a 3.

Next is, what kind of shooter is he from distance when he is on the move instead of a spot up? He shot an air ball and didn't look very comfortable doing it - with the game on the line.

The next thing that jumps out at me is "Level of Significance." Remember, reliable and valid, per 100 possessions. Over the course of an entire game, you might be able to win a lot of games with players that do that. Over the course of the entire season, you can become pretty consistent. You know what to expect, game to game. That's what analytics are for. Per 100 possessions! You coach by the analytic, it paints you a picture, you mold your team, and you develop a style of play. Per 100 possessions!

In this situation, this was ONE possession. Not 100 possessions, ONE possession. I don't care what the analytics say. I don't care for 3 or lay-up only. The question you need to have answered is, "With the game on the line, do you really want a 23% shooter with the shot that decides whether you win or lose?"

Are Analytics Leading Us Down a Bad Road In Some Cases?

I think this is one of the areas that I think that analytics is leading us down a bad road.

All parts of the game are not the same. The end of the game is different from the start of the game.

The analytics are not the same at all levels. At the HS level, and certainly the youth level, you don't have enough data to really have a "reliable and valid" set of statistics.

All players aren't the same. The analytics for each player is going to be different. If you are analytically driven, you need to have ALL the information.

If not, it may come up to bite you at a bad time. As a prisoner of analytics, you try to force the game into your stats. That's what happened here.

Now, again, it is not my intention to criticize the coach or his decisions. Quite the contrary, I think he's a terrific coach. His preparation and decisions allowed his team to go down to the wire against a better team. This is just an example that illustrates what I think is a decline in situational coaching. I see it everywhere.

What determines winning and losing is control of critical situations.

When the critical situation occurs at the end of the game, is your team prepared? I don't mean what play are you going to run. There's more. Do your players have a grasp of the situation? Do they understand the change in mentality between this situation and others? Are they clear on the objectives for this one possession game?

Other Resources From Don Kelbick

The Attack & Counter Skill Development System

The Simplified Motion Offense

Continuity Zone Offense & Concepts to Beat Any Zone Defense

Match Up Zone Defense

What do you think? Let us know by leaving your comments, suggestions, and questions...

Comments Most Likes First   Oldest First   Newest First Kenny Williams says: 4/25/2020 at 12:32:09 PM Hello, Coach K always has good thoughts and pointers in his articles and film productions. I would like for Coach to better flesh out his assertion that “foul shots are devalued.” In fact, most analytic articles I have read and coaches I know highly value the FT. Thanks.... Like   1 reply   login   register Don Kelbick says: 4/27/2020 at 8:13:16 AM See my comment to Aaron below. Like     login   register Dom says: 4/25/2020 at 11:23:49 AM Great article, and great points. Data, used appropriately, is a weapon for coaches. But if you are unable to gather appropriate data, then the findings are meaningless. In the article's example, the same shot by a different player, or a different shot by the same player, might have been better. So much goes into it. I run analytics and coach for a competitive high school team, and the stats we use to dictate strategy are FAR different than what NBA teams use. And it all depends on what data is statistically significant for our team. Basically, What data can we trust? Used the right way, analytics are helpful. Used inappropriately, you might see ghosts and not realize it. Like     login   register Paul says: 4/25/2020 at 10:46:39 AM Morning. Great article. As a former youth b-ball coach... I agree fully with Coach K. Two comments: (1). The stats that us coaches see in games is completely different from the stats we see in practice. Separation of these by coaches in developing their pre game plan and adjusting their in game plans is critical. I have seen this countless times. (2). I believe Coach K hinted to this in his article. The glory of the 3 pointer and the level of greatness assigned to it by watching the likes of Curry, etc (and I love watching these guys hang the 3).... has done enormously more harm to youth b-ball then anything else. I have seen youth b-ball programs from grades 5 to 8 that allow their players both in practice and in games to attempt a 3 point shot even though the kids are not strong enough due to age, lack the coordination at the young age and lack the shooting mechanics to even set them up for a higher success shot. Precious time is wasted allowing this glory 3 to be made INSTEAD of teaching the BASICS of b-ball! Like Coach K said, the mid range shot is dead....due to the focus on the glory 3 instead of in the lane b-ball skills. As former 5-6 grade coach, 15 minutes into the first team practice and seeing every kid stand at the 3 pt line launching wild and undisciplined shoots I called the team in and made the decision to band 3 pt shots. Taking them, regardless if they were made or not or even won a close game for us would result in benching and/or a lot of laps/pushups. I explained this to the parents which were in agreement. I wanted to focus the season, at that age, to basic b-ball skills development. What I saw in the season were kids working hard, developing, gaining confidence in the lane and confidence in each other.........we went undefeated that year....because of the focus on the basics and not the glory 3. Like     login   register John Arcaro says: 4/25/2020 at 10:09:58 AM Yes, MANY of the coaches under 40 have turned their back on the midrange game. When considering the 3-point shot, it is important to remember that the average NBA player is SIGNIFICANTLY bigger, stronger, and more athletically developed than the average HS player. Like     login   register RT says: 4/25/2020 at 9:57:15 AM Although there’s “art & science,” and the alchemy of everything else, eg, experience, how a player reacts to stress, etc., I’m surprised that this article proves the other side of the argument. First a correction, FTs are the 2nd best PPP shot behind shots around the rim. (But you may foul on a 2 for 1 end of period possession series to control WHO shoots (bad FT shooter) and to preserve clock for your 2nd possession, etc.). Second, Coach Kelbick (my son attended Attack & Counter camp, I have the DVDs) proves the counter point: the Expected Value for a 23% shooter from 3 is .69. That’s horrible and it’s based on averaging open shots, disrupted shots, and this shot is for the win. No analytics coach deliberately chooses a shot w less than 1.0 expected value for his/her own team. That’s what you’d do in the above 2 for 1 scenario where you use analytics to CHOOSE who on the other team shoots. I don’t think when you’re at home you play for the win vs OT. Based on analytics, you get that guy a shot inside. Same conclusion. That said, analytics don’t always result in wins. Players still have to execute. Coaches still have to coach situationally. But if the process fails and the result isn’t achieved it doesn’t mean the analytics shouldn’t be used. They’re an input to decision making and as teams choose to shun the midrange, if defenses adjust, it opens what? The midrange. But Expected Value for a team’s midrange (leave out individual for now) is going to around 1 in close mid-, .9 in deep mid- inside the arc, higher from 3, highest from 3 corner (of all arc shots), higher from FT and highest around the rim. Sorry for any typos. Like     login   register Aaron says: 4/25/2020 at 9:26:44 AM Good article, but I do have a few issues with it. First, foul shots are VERY analytically friendly. See James Harden, the most 'modern' efficient scorer. And points per possession is meant to be pace-adjusted, so it doesn't matter if you have 50 possessions or 150. I do agree that each possession is different, and that's where situational awareness of what shot is good in that moment is important Like   1 reply   login   register Don Kelbick says: 4/26/2020 at 3:15:08 PM Aaron, I think you misinterpreted my comment about foul shots. Foul shots are second, in the NBA, to layups (different than shots at the rim). I probably expressed the thought poorly. Harden had a game last year where he shot 2-9 from the floor and still scored 29 points. Those points had to come from somewhere. What I meant by de-valuating the foul shot is this. A major part of the game used to be getting into the bonus as early as possible. Players would make significant careers out of their ability to get to the foul line. Two players that come to mind quickly are Adrian Dantley and Mark Aquirre. Both had good careers. Dantley was good enough to be in the Hall of Fame. At 6-5, he played in the post. It was a big part of Charles Barkley's career as well. Without their ability to get to the foul line, their careers probably would not have been so significant. There doesn't seem to be room in the game today for players like that because the ability to get a foul shot does not seem to have the same value as it once did. Just as an example, we were playing one of our many games against the #1 team in the country (we played in a great league) and they had a post player who would become a 2x National Player of the Year. We had a great game plan to limit his touches and for much of the game, we executed it very well. Ater the game, which we lost frustratingly closely, my assistant said to me, "We executed the game plan well and really limited the post player's affect on the game." My answer was, "Did we really?" We went to watch the film and it showed that, at the start of the 2nd half, they put him on the block and went to him on 8 straight possessions. He only scored on 2 of them and was not a factor for the rest of the game. However, he created 7 fouls in that period and they were in the bonus for the last 15 minutes of the game. No matter what we did, we could not overcome the advantage they created by going to the foul line so many times more than we did because of that. Today, I don't think that is in too many game plans. That being said, please don't lose sight of what the article was meant to be about. It's not important if a metric is adjusted for pace or not. We have 1 possession to win or lose a game. Is the emphasis on the larger metrics hiding the value of individual possessions and leading us to make decisions we wouldn't otherwise make? Just food for thought. Like     login   register

Leave a Comment Name : Email (not published) : Four times three is equal to?  (Prevents Spam) Answer :  Load New Question Comments : Leave this Blank :