Archive for the 'Projections' Category

In the comments section of our Eastern Conference Preview, a reader asked for the minutes played distribution that was used for our projections. Please see the complete table after the jump.

Read the rest of this entry »

Click here for the Western Conference Preview

After many months of projections, it's finally time to unveil our official preseason forecast. They're based on our Win Shares projection system, with rosters/depth charts current as of Sunday, October 25, 2009. We used Win Shares to establish a base team strength rating for each squad, and then plugged those into the handy log5 formula and ran another 10,000-season simulation. What follows are the rosters (newcomers are highlighted in yellow), projected per-minute stats from our Simple Projection System, projected WS, the results of our 10,000-season sim, and my brief (200-words-or-fewer) thoughts on each team. So let's start the show, in reverse order, beginning with…

Read the rest of this entry »

Yesterday I made some changes to the Simple Projection System (SPS). The basic structure remains the same (it is supposed to be simple, after all) with the following modifications:

• The weights for the previous three seasons have changed from 5-4-3 to 6-3-1.
• The regression to the mean component now uses 1000 minutes of league average performance rather than 5000 minutes.
• The age adjustment is now (28 - age) * 0.004 for players younger than 28 and (28 - age) * 0.002 for players older than 28.

Read the rest of this entry »

This week, the news Knicks fans have been hoping to hear finally came: David Lee and Nate Robinson re-signed with the club, each inking 1-year deals in the $5-7 million range that (most importantly) will not interfere with the Knicks' long-awaited free agent pursuit of LeBron James, Dwyane Wade, Chris Bosh, or any of the other headliners in the star-studded FA class of 2010.

Read the rest of this entry »

So, after a long week of programming, here's the first pass at a Win Shares-based projection system... The calculation is structured the same as the Simple Projection System, except that this method is used. Playing time was estimated based on Ed Kupfer's formula for game missed due to injury, a minutes per game formula (MPG = 10.16 + (0.557*MPG_Y-1)+(0.029*MPG_Y-2)), and total team minutes were forced to reconcile to 60% for starters, 30% for 2nd-stringers, and 10% for benchwarmers (HoopsHype's depth charts were used for the designations). And a final note, remember that these projections will be unrealistically close to the mean for players, teams, and playing time numbers, because of the heavy degree of regression being used. Anyway, here are the results:

Read the rest of this entry »

Recently, we've been focused on projecting various "meta-stats" for the upcoming 2009-2010 NBA season, starting with Statistical +/- and now moving to Win Shares, with the goal of predicting how each team will finish based on their current rosters. Rate stats are, of course, relatively easy to project for basketball players because they don't fluctuate a ridiculous amount from season to season -- they are generally a function of the player's ability level, with teammates, coaches, and other external factors playing a role as well. Playing time, however, is always difficult to project because much of it stems from factors outside the player's control, like injuries, coaching decisions, trades, etc.

Read the rest of this entry »

This tool has been live on the site since the preseason of 2008, but I'm not sure many people are aware of it yet... It's called the Simple Projection System, and it's a pretty unique feature as far as basketball projections on the internet are concerned. Plenty of other sites have projections, of course, but most of them are either completely non-scientific (read: "guesswork") or based on heavily-guarded methods so secretive and requiring so much proprietary data that no layman could possibly hope to recreate them for themselves. The SPS, though, was borne out of the same spirit that had led sabermetrician Tom Tango to create the Marcel projection system for baseball, so named because it was simple enough that a monkey (in this case, Ross Geller's pet monkey on Friends) could replicate its results. Similarly, our SPS at Basketball-Reference doesn't need to use fancy similarity scores or umpteen-thousand obscure variables in order to spit out a series of projected per-minute rates for every player who played a game the year before. Instead, it simply uses past results, a heavy regression to the mean, and a simple aging adjustment, creating surprisingly credible results with this no-frills approach.

Read the rest of this entry »

Last week, we took a very preliminary look at what our Statistical Plus/Minus projection system saw in the cards for the 2009-10 NBA Season. To project minutes played, we used a very simplistic regression equation that took a weighted average of a player's minutes over the past three seasons and regressed it heavily to the mean. Of course, this is a very rough way to estimate what a player's minutes will be next season; in fact, the standard error of the playing time regression (done on all players from 1978-2009) was 674.8, meaning that the prediction was likely to be off by a significant amount in either direction, too high or too low. As you might guess, this could severely impact the accuracy of the projected standings.

Read the rest of this entry »

Since everyone seems to be jumping on the projection bandwagon right now, here are the preliminary projected standings for 2010 (I'll explain the process at the bottom):

Read the rest of this entry »

Following up on our post from Monday, where we looked at the % of minutes played by common players between teams in back-to-back years, here's a list that uses the same methodology -- but instead of minutes played, we're looking for the smallest % of common individual possessions used (as defined by Dean Oliver in basketball on Paper) between players on a team in consecutive seasons:

Read the rest of this entry »