Complete Halftime Analysis

Most halftime shows include comments from former players and coaches about how teams must “pick up the intensity” or “step up” in the second half. In basketball games, analysts talk about strategies like “getting the ball inside” or “taking better care of the basketball.”

While such input from former players and coaches is valuable, it is far from complete analysis. You need to get into the numbers for that to happen. Sports analytics provides powerful insight and doesn’t need to involve anything complex. The concept of regression to the mean reveals far more about what to expect in the second half, and only requires comparing a few key stats.

Let’s say one team shoots 8-for-10 from three-point range to grab a 10-point lead while the other clanks 1-for-9. If both teams entered the contest shooting 40 percent on the season, it is almost certain that both teams regress to the mean – the 40 percent mark – in the second half. Defense impacts these stats, but randomness (or luck) is a huge factor in small samples. As more shots get taken, teams should move toward their season percentage and cause the score to tighten up.

Free throw percentages are another great stat to examine at halftime. Unlike three-pointers, they aren’t defended. So randomness plays an even greater role.

I believe broadcasts will soon have a statistical expert offering such insight – alongside coaches and/or former players – on halftime shows. But until that happens, we can all do it ourselves.

Do-It-Yourself Three-Point Shooters

With five seconds left on the shot clock and all their teammates covered, guards have to create. Some perform much better at it than others.

Creating and then converting a three-pointer is basketball’s version of a grand slam. A 24-second violation nets nothing – and a wild shot attempt isn’t much better. So it’s a three-point swing if a player can nail a three off the dribble. Since few players shoot high percentages in these situations, the players who excel have tremendous value.

After evaluating all three-point shooters for volume, accuracy and the ability to create their shot without an assist, Spurs guard Gary Neal stood out. He hit 41.9 percent overall from three-point range last season, even though only 54.2 percent of his made threes were assisted. On average, 84.2 of NBA three-pointers were assisted in 2011-12.

Most top three-point marksmen have a very high percentage of their threes assisted. That’s no problem of course, it’s their job to spot up and drain threes. But it makes players who can convert threes off the dribble even more valuable, especially for teams that don’t get many open three-point looks from their set offense.

In addition to Neal, other players who shine in this area include Kyrie Irving, Kyle Lowry, Jose Juan Barea and Lou Williams. 

Time to Move On

If a point guard had these averages after his name (3.6 PPG, 1.9 APG), most people wouldn’t think much of him. But if you change them to 21.3 and 11.3, that’s a different story. These stats are actually for the same player during the same timeframe. They show Jeremy Lin’s production in his first nine games this season before he saw regular action and exploded on the scene. The second set of numbers reveal Lin’s numbers per 36 minutes of game action. He began his strong performance before Linsanity, although few noticed.

Once some ideas become entrenched in sports, it’s difficult to change. I have no idea why basketball started using per game statistics, but it’s time to stop. NBA players – even those that we call regulars – vary widely in minutes played per game. We already saw how useless they are for bench players. Per game stats cause misconceptions about performance, and hamper the ability to identify breakout players. We should evaluate players per minute or, better yet, per possession. But the latter concept is too great a leap for the mainstream.

In the meantime, here’s an alternative to points per game. True Scoring Rate (TSR) is simply a player’s points scored per 36 minutes. Why 36 minutes? That’s the approximate average for a top NBA starter. This way we find out who scores at the greatest rate, something points per game has never done.

Points per game shortchanges players like Kyrie Irving, who had played just 31 minutes per game (through February 23). Irving ranked 23rd in points per game (18.1), but his 21.1 TSR placed 15th, ahead of point guards Tony Parker (20.3) and Chris Paul (19.1). Both surpassed him in points per game, largely because they saw more minutes.

True Scoring Rate is not really new, just a way of simplifying per minute rates. It’s easy to understand and calculate. Most importantly, it adjusts for the large discrepancies in playing time.

The next Jeremy Lin might be sitting on an NBA bench right now, and TSR can help find him.

Keeping it Simple

I have been reading “The House Advantage” by Jeff Ma. The book explains the importance of incorporating data into the decision process. Ma uses examples from his background in sports and as a member of the MIT blackjack team; he was the central character in the best-selling book “Bringing Down the House”.

In one chapter, Ma writes about the approach he takes when consulting with businesses, whether sports-related or not. He summarizes it without using any complex statistical terms, but as taking information from the past to make decisions about the future. Whether in sports or business, that process creates an edge.

This applies to most sports decisions, whether for free agency or the draft. What trends are in a player’s past that bode well for their future? That track record may be long and consistent with plenty of data, as with Albert Pujols or Tim Duncan. Or it could be brief, such as the case of a young lefty relief specialist or a college freshman entering the NBA Draft. Generally speaking, the more the data, the easier it is to project performance going forward.

The quality of the data is also important. When projecting Ubaldo Jimenez’s performance for the season’s final two months, his impressive won-lost record through July is not all that helpful. His ERA is more valuable, but still not the best predictor. The best projections come from examining Jimenez’s walks and home runs allowed, the percentage of batters he has struck out, and factors like left on base percentage and batting average on balls in play. Performance in previous seasons matters too, as that is all part of his track record.