Dre Alvarez (@nerdnumbers) is a Co-Editor for the Wages of Wins Network and is also in charge of handling the stats data. He’s a long time fan of Colorado Sports, depending on the weather. He’s an even bigger fan of the stats, data and all things nerdy.
Lessons in being a GM
Today I’ll be giving a lesson in how to be a GM in not one, but two sports. The NBA and NHL both have their ideas about how to improve a team. I’ll address the notion that an NBA team needs a superstar to contend and the theory that an NHL team needs a top goalie.
You Need a Star to Contend in the NBA
In the NBA everyone knows to be a contender your team needs a star. This thinking is actually quite accurate. I’ll define a star as a top 25 player in the NBA using the Wins Produced metric. As a GM if you knew all the stats of last season could you find a star? Let’s take a look at the top 25 stars of this season with their last season rank and numbers for perspective.
| Team | 2011 Rank | 2011 WP | 2010 Rank | 2010 WP | |
| Kevin Love | 1 | 25.3 | 22 | 12.1 | |
| Dwight Howard | 2 | 24.3 | 2 | 22.3 | |
| LeBron James | 3 | 22.6 | 1 | 27.2 | |
| Chris Paul | 4 | 20.8 | 24 | 11.5 | |
| Dwyane Wade | 5 | 18.1 | 7 | 17.8 | |
| Pau Gasol | 6 | 17.0 | 12 | 15.4 | |
| Zach Randolph | 7 | 16.8 | 17 | 14.3 | |
| Blake Griffin | 8 | 15.5 | N/A | N/A | |
| Steve Nash | 9 | 14.5 | 13 | 15.4 | |
| Kris Humphries | 10 | 14.3 | 141 | 3.4 | |
| Kevin Garnett | 11 | 14.2 | 28 | 10.1 | |
| Al Horford | 12 | 14.2 | 21 | 12.7 | |
| Paul Pierce | 13 | 13.8 | 52 | 8 | |
| Lamar Odom | 14 | 13.6 | 14 | 14.6 | |
| Kevin Durant | 15 | 13.5 | 3 | 19.7 | |
| Landry Fields | 16 | 13.3 | N/A | N/A | |
| Rajon Rondo | 17 | 13.2 | 8 | 17 | |
| Jason Kidd | 18 | 13.1 | 4 | 19.6 | |
| Tim Duncan | 19 | 13.0 | 10 | 16.2 | |
| Andre Iguodala | 20 | 12.8 | 15 | 14.4 | |
| Tyson Chandler | 21 | 12.2 | 178 | 2.3 | |
| Derrick Rose | 22 | 12.0 | 78 | 5.7 | |
| Manu Ginobili | 23 | 11.7 | 20 | 13.4 | |
| Russell Westbrook | 24 | 11.6 | 43 | 8.7 | |
| Kobe Bryant | 25 | 11.6 | 32 | 9.9 |
Table 1: 2011 Top 25 NBA Players using Wins Produced and their 2010 numbers.
Taking a look at these numbers here’s some strategies I’d offer to NBA GMs to acquire a star.
Strategy 1) Have and Keep a Top 25 Player
Good Players Stay Good
- Success rate in 2011 for top 25 player: 15/25
Yup, if you have a great player you should keep them. They’ll probably be great next season. Only seven returning players on our list weren’t in the top 25 last season and only LeBron actually left his team for another. Not only do the top players stay at the top, their play seems to stay pretty consistent. Most of the players on this list stayed within 5 Wins Produced of last year’s totals.
Strategy 2) Have a Former Top 25 Player that had some injury problems and hope they get better
Out of Street Clothes and back at the top.
- Success rate in 2011 for top 25 player: 4/25
NBA players — when they recover from injury — can revert to form, and if you had a formerly great player they can get back there. Kevin Garnett, Paul Pierce, Kobe Bryant and Chris Paul fall in this category. In the last three years each of these players has been a top 25 player in the league. Last year hindered their performance (it still didn’t keep Paul out of the top 25 though!) but this year they returned with a vengeance.
Strategy 4) Have a talented young player mature into a star
Young and improving.
- Success rate in 2011 for top 25 player: 3/25
Young players improve. If you’ve got a talented player that’s been improving next year they make finally jump into the top level. Derrick Rose, Kevin Love and Russel Westbrook all saw a marked improvement this season and for the most part saw a jump in team performance. (2 out of 3 aint bad)
Strategy 4) Get a top Rookie
Can you imagine it?
- Success Rate in 2011 for top 25 player: 2/25
Strategy 5) Get a Top 25 Player
South Beach wins!
- Success rate in 2011 for top 25 player: 1/25
This is a great strategy but good luck getting teams to give up their stars. Only LeBron James falls in this category. Turns out Miami’s decision to get him worked out well. The 76ers are rumored to be shopping Andre Iguodala. Other GMs should take note.
Strategy 6) Find a former top 25 player that has recovered from injuries.
<img class=”size-medium wp-image-4360″ title=”
- Success rate in 2011 for top 25 player: 1/25
Tyson Chandler in 2008 was ranked 17th with 14.1 Wins Produced. In 2009 he got injured and has been bounced around since. This year he returned to form and helped the Mavericks win a finals. A risky move, but it can work.
Strategy 7) Get an above average player that has been played limited minutes and hope they reach the next level
Do Women and Clothes make the NBA player?
- Success rate in 2011 for top 25 player: 1/25
I wouldn’t rank this as a great strategy but sometimes teams have good players that have been cast as “role players” and are given limited minutes. It’s not the biggest stretch to assume if they play well in limited minutes, they might play well with starter minutes. You might also get extremely lucky and the player will play amazingly.
General Notes
The idea that these players are needed for success in the NBA is very accurate. A case in point is every team that made it to the 2nd round of the playoffs last season had at least one. The hard part is getting one of these players. Only Miami, New Jersey and Dallas were able to grab one in the last offseason and we can see the fates of two of those franchises turned around immediately. The problem is that there are over 500 players in the NBA. and with only a small handful of great players, it’s simply not possible for most teams to hope of getting them. This gets even harder when we notice multiple teams (Boston, Dallas, Los Angeles, Miami, Oklahoma City and San Antonio) grabbing more than one. In short, when your GM tells you they’re going after a star to contend next season you should be happy. Just don’t be optimistic that they’ll actually get one.
You need a top Goalie to Contend in the NHL
Recently we reviewed the great work of Stacey Brook and Dave Berri about NHL goalies. Here are the important lessons from this study.
- NHL goalies are for the most part the same. The difference between a great goalie and an average goalie is small. And even the best goalies aren’t huge difference makers. The best goalie will earn you around an extra five wins and the worst may cost you around three.
- Goalies are inconsistent year to year.
If your GM is spending the offseason in the front office doing their best to acquire last year’s top goalie, I wouldn’t be that thrilled. For fun though let’s redo our NBA exercise and ask our GM in 2011 to acquire a top 10 goalie based on the 2010 season using the Wins Above Average (WAA) metric (save% and shots on goal in terms of wins).
| Player | 2011 Rank | 2011 WAA | 2010 Rank | 2010 WAA | |
| Tim Thomas | 1 | 4.75 | 19 | 0.67 | |
| Jonas Hiller | 2 | 2.74 | 10 | 2.01 | |
| Ondrej Pavelec | 3 | 2.36 | 66 | -1.07 | |
| Pekka Rinne | 4 | 2.10 | 38 | -0.04 | |
| Henrik Lundqvist | 5 | 1.99 | 6 | 3.11 | |
| Roberto Luongo | 6 | 1.88 | 21 | 0.45 | |
| Marc-Andre Fleury | 7 | 1.80 | 73 | -1.66 | |
| Tomas Vokoun | 8 | 1.71 | 2 | 4.26 | |
| Carey Price | 9 | 1.59 | 26 | 0.21 | |
| Cam Ward | 10 | 1.57 | 15 | 0.93 |
Table 2: 2011 top 10 NHL goalies with 2010 numbers.
The ideal strategy
Just pick one, doesn't matter.
I don’t need multiple bullet points and examples to explain the ideal strategy here. It’s actually pretty simple.
- If you’re an NHL team and you want a top 10 goalie: Have a goalie on your team. Play them. Hope they play like a top 10 goalie.
Yes, that is about the best strategy you can adopt.
One should also note — as Dave and Stacey noted in their paper — that performance of goalies seems to depend on the defense playing in front of the goalie (i.e. goalies depend on their teammates). So maybe if your goalie isn’t playing well maybe you should take a closer look at your defense.
We repeat ourselves: Goalies are hard to predict.
Only three goalies in the top 10 from last season return to this year’s top 10. Even then it’s not that great. Lundqvist and Voukun both saw massive drops offs in performance. Hiller saw a slight increase. Tim Thomas, this season’s top goalie, was barely better than average last season. What’s more the number of terrible goalies from last season (Pavelec, Rinne and Fluery) that are top ten matches the number of great goalies that returned top the top ten! In terms of making the right choice it’s next to impossible. Compare this with the NBA where a top player regardless of position is almost a lock.
Being able to predict a top goalie has not been shown to be easy. That isn’t to say a top goalie doesn’t help. Tim Thomas and the Bruins did win a Stanley cup and Thomas’ regular season help amounted to around ten points in the standings (the difference between 3rd place and 8th place in the Eastern Conference). But can the Bruins count on him leading them to greatness again next season? Should other teams pursue him like crazy and hope he throws a press conference saying he’s bringing his talents to their town? No. The best strategy might just be to put the same goalie back in net, even if they played badly last season, and hope they play like a top 10 goalie.
-Dre
Wages of Wins Journal 
Nicholas Yee
July 22, 2011
The article tends to take an overly random tone to goal prevention in the NHL, but from what I’m understanding the real key is to spend money on defencemen.Now my question is if there is any consistency between a defenceman’s play from season to season? Is there much value in a offensive defencemen? Or does it relate to basketball sort of like a non-rebounding big man?
Could this goalie thing, be a big reason for the high amount of NHL upsets? In comparison to the NBA in particular. Or it every other major North American sport prone to upsets because of small sample size? I mean, though basketball is only a 7 game series at most, each game there are 100’s of possessions, and points scored. While in hockey, soccer, or football, the number of scoring opportunities is much lower, so 1 random play, can decide a game if it results in a goal or TD, because scoring is so scarse. While in basketball, one random play will at most result in 3 points.
Dre
July 22, 2011
Nicholas,
I agree on the goalies. I will just say as my perspective as an Avs fan that EVERY year we hear how we need a good goalie to compete. You are right that good defenses have a huge impact on shots on goal and as a result goals scored and wins.
I agree with you in part on the upsets. A high shooting game would be around 60-70 shots total. Most of these aren’t “Real” shots. So in essence you have two teams maybe with 10 “real” attempts each at max and luck can influence the winner.
Another point is Stephen Jay Gould’s point on reaching an athletic peak. I don’t think the difference between the best players and average players in the league is that great. In the NBA LeBron is 3-4x better than a normal player. In the NHL I might say 2x better would be optimistic and even then their minutes are limited (<half the game)