This story has already received a fair amount of attention at ESPN. Marc Stein at ESPN.com penned a column last week examining the plethora of players who did not play last season but hope to play in 2007-08. And then this morning, Kelly Dwyer – subbing for Henry Abbott – also noted the number of active seniors hoping to re-join the Association. At the end of Dwyer’s column was a link to an ESPN Insiders entry posted by John Hollinger, which covered similar ground.
So the basic story has been touched upon. Still, I am going to make a small effort to add something worthwhile to this discussion.
My focus is going to be on two of the most prominent players on the list – Reggie Miller and Allan Houston. Currently each is an NBA analyst, Miller with TNT and Houston with ESPN. The similarity between these two extends beyond their current occupation. Each player was an NBA shooting guard who was known for his outside shooting. Miller leads all players in NBA history with 2,560 three point shots made. Houston is currently 11th on this list, but actually converted his shots from beyond the arc at a higher rate than Miller. Houston made 40.19% of his three point attempts while Miller converted only 39.47%. Okay, not much of a difference. Still, technically Houston was more efficient from downtown. And given that Houston is six years younger than Miller, it looks like Houston would be the better bet in 2007-08.
The problem with this assessment is that basketball is not just about launching bombs. And when we look at all that these players did, we see that Houston was not nearly as effective as Reggie Miller.
Like the discussion of Pistol Pete, it’s useful to compare each player to what we see from an average shooting guard.
Table One: The Average NBA Performance of Reggie Miller and Allan Houston
The temptation with each player is to focus on scoring, but I think it’s useful to focus a bit of attention on what each player did with respect to the non-scoring aspects of the game. In terms of rebounds, steals, assists, and blocked shots, both Miller and Houston were below average. With respect to personal fouls each was above average. Finally when we look at turnovers we see that Miller was above average (meaning he did not turn the ball over as often as you would expect), but Houston was again below average for an NBA shooting guard.
Okay, what do these differences mean? Let’s first focus on Houston. Per 48 minutes, Houston cost his teams 1.5 rebounds, 0.8 steals, 1.3 assists, and 0.2 blocked shots. He also committed 0.2 more turnovers. Although he committed 0.2 fewer personal fouls, overall his non-scoring efforts did not help much. To see how much, consider that each rebound, steal, and turnover are worth – in absolute terms – about 0.033 wins. Assists, blocked shots, and personal fouls – again in absolute terms – are worth about half that amount. Given these numbers, Houston’s discrepancies with respect to the non-scoring elements of the box score cost his teams about six wins per 82 games (given that Houston played about 34 minutes per game in his career).
Repeating this analysis for Miller reveals that his difficulties with respect to the non-scoring elements cost his team about 2.2 wins per 82 games (again, given that Miller also averaged about 34 minutes per game in his career).
Okay, so we see when each player wasn’t shooting he wasn’t exactly helping. And Houston was especially a problem with respect to the non-scoring aspects of the game.
Of course, when it came to shooting, both players were an asset to their teams. The average shooting guard scores 0.96 points per field goal attempt [(PTS-FTM)/FGA]. Houston scored 1.00 points per shot from the field, so he was certainly above average. Given that he took 20.4 shots per 48 minutes, Houston’s ability to score at an above average rate added 0.7 points per 48 minutes. And what are these points worth to an NBA team? About 1.7 wins. That’s it.
Now let’s turn to Miller. Miller scored 1.09 points per field goal attempt in his career. He took 17.6 shots per 48 minutes, and hence he scored 2.2 points more per 48 minutes than you would have expected from an average shooting guard. These 2.2 points are worth an additional 4.3 wins over an 82 game season. Miller was also able to get to the free throw line and convert at higher rate than the average guard. And this ability also created wins for his team.
If we put it all together, we see that Miller was able to overcome his deficiencies with his shooting. Houston, though, was not.
Table Two: The Season-by-Season Performances of Reggie Miller and Allan Houston
When we look over the career numbers for each with respect to both Wins Produced and Wins Produced per 48 minutes [WP48], we see what Miller and Houston meant for their respective teams. In every season Miller played, he was above average. In every season Houston played, he was below average.
And that is another feature of NBA performance that I often emphasize. Year after year we see that players who are well above average stay above average. Players that are below average remain below average.
So what can teams expect if they sign Houston or Miller? Both are now older and age does make a difference. Given that Houston has never been above average, we can expect an older Houston to still be a problem. Miller, though, has always been above average. But one wonders if that’s going to be true at 42 years of age.
In reading over the analysis of Stein, Dwyer, and Hollinger the same story was told. And I think their conclusion would also be reached by any NBA analysts, including Houston and Miller. It’s probably a good idea that teams let each player continue in their role as NBA talking-heads. A team taking a shot with either of these senior citizens can expect such a shot to catch nothing but hot air.
– DJ
Wages of Wins Journal 
Gareth Lewin
August 11, 2007
I think this is one of those cases where the stats don’t say everything.
I’m a huge Reggie fan, he’s my favorite NBA player of all times. I agree with you that at 42 he shouldn’t return to the game.
I also agree that he probably wouldn’t help a team directly, probably would even be below average.
But Reggie was always more than just a player. He was a real leader. Atleast that is what it seemed like thru the filter that is television.
I can see a team with mostly young player benefiting from a Reggie in their locker room.
That said, I really hope he doesn’t try to come back, for his own good.
Gareth Lewin
August 11, 2007
Oh one comment I need to add. I hadn’t read where Reggie was thinking of going. Now I see it’s Boston.
That is just a bad idea to me. He wouldn’t be able to be a leader in Boston, and he could only be doing it with the hope of winning a ring (Reggie deserves one).
Since I doubt that Boston are getting anywhere near a ring (Chicago will beat them in the playoffs no matter what stage they meet) Reggie is making a mistake.
But on the plus side personally, whereever Reggie goes, my heart goes. So at least I’ll get to support KG :)
Paul
August 11, 2007
These stats show that shooters get too much glory and pay, but in the right team context a shooter only can work. Houston helped make Knicks 11th best shooting team in 2000-1, the last year of his previous deal. Van Gundy and the rest of team contributed the 3rd best team defense. That combo was powerful. Houston wasnt the only good shooter that year but he took the most shots and maybe helped allow others to stay in a usage level where thy could succeed and helping to space floor for the bigs to operate and keeping Glen Rice in single coverage. Larry Johnson and Glen Rice left summer 2001 and they re-signed Houston to a huge deal that he didnt really deserve to try to hold onto something of a successful run. A bad move but not atypical. I thought they overpaid at the time. The final major blow to the previously fairly successful team with guys filling different roles was Van Gundy quitting. Without the defense the Knicks werent good enough elsewhere. With strong defense Houston fit and helped. Without it he didnt because he wasn’t good on the other parts of offense or defense. The individual stats tell a lot of the story, but not all of it. Small variations from position norms can sometimes be covered without harm by teammates or other team strengths. GMs should know and give weight to the individual stats but they have to know their team context. Ultimately teams win. Players contribute to wins and losses in rough accordance with individual stats but individual stats do not fully and completely accurately capture a player’s win contribution to me because each context is unique and dynamic. Knicks lost on Houston signing because the context in which he was ok as lead scorer changed dramatically for the worse. Some of that could have been forseen with player departures, Van Gundy has supposed to be the bulwark and he walked away. Houston’s later injuries made the situation even worse. Knicks made many mistakes and got a couple big hits of bad luck. Overvaluing a shooter was a mistake but they had made it work before. To win championship you can’t make more than 1 -2 bad calls on personnel. but you can win 40-50 with an imperfect roster. Getting an ideal roster is quite hard.
Paul
August 11, 2007
Similarly Reggie helped balance the floor for Indy’s inside game and the defense there played a prominent role in team success and Reggie’s presence allowed Pacers to play more defensive players. You stray from individual stats norms at some peril but if you can produce net team edge at the sum of the five positions, two ends of court and the various parts of the game in the end, then you have started the journey of being above average.
I guess a question could be does it make sense to strive equally hard for advantage at everything or to target certain parts of the game to try to dominate and make them your edge producing engines. It depends on your roster (or available options to change the roster), your coaches strengths and the nature of the league at the time and particularly your conference and division and how much you have achieved so far and what your realistic goal is for the current season. Ultimately I am a strong believer that teams who are strong throughoout the 4 Factors with no more than 1 or 2 weaknesses are best. So it is better if your offensive main guys are complete players- more like Ginobli, Duncan and Parker or the Pistons’ main guys.
Alex
August 11, 2007
I don’t understand your conclusion that Reggie Miller probably wouldn’t be a productive player, as it’s based solely on the fact that he’s 2 years older than when he last played… Considering the Celtics are only courting him to be a 10-15 min/game bench player, and since such players are traditionally below average, isn’t it possible that Miller could play slightly worse than his previous productivity level… and yet still be the best available option as a limited time backup to Ray Allen? (Or, put another way, if they shouldn’t sign Reggie to be their backup SG, who should they sign, with only the veteran minimum available and no trade assets)