Pages

Fantasy Notes

10/18/2007
As a service to all my loyal blog readers, I've put together updated preseason stats in a spreadsheet at th e following link:

http://www.coltonindex.com/pre0708.xls

A couple of things worth noting from the preseason numbers so far:

- Dwight Howard has been a beast. The biggest thing that sticks out is his 80% free-throw percentage. obviously, we don't want to jump the gun - we're talking only 25 attempts here - but this is definitely worth keeping an eye on. Howard is generally still vastly overrated relative to his average draft position, especially in 9-cat leagues. In my 8-cat drafts so far, e's gone in the first round in each, including fifth overall in a draft last night. Usually, that's a sure fire death sentence and an indicator that the guy doesn't have a clue. Now there is a greater chance that it ends up working out. I avoided amare like the plagued last year and got burned. This year, I wouldn't even consider picking howard until 17 or 18 which means I won't get him at all. Somebody's always going to overpay for him. We'll see how it shakes out.

Hypothethically, if howard were a 80% shooter, he'd be #1 or #2 overall in 8cat leagues and in the top 10 or 12 in 9 cat leagues. What's more difficult to quantify is the impact that would have on other categories. Howard's ft attempts shot up last year as opponents figured out that it was one of the only ways to stop the man child. What if that was no longer a viable option?

In my projections, I haven't gone overboard with howard's numbers. I've bumped him up from 60.8% to 61.4% for the time being. I will continue to monitor him closely.

- Grant Hill a three-point gunner? Going to the Suns does wonders for one's fantasy prospects. I think I could be a borderline fantasy-worthy player if I played 30 minutes a game with Steve Nash. Hill has hit a total of 9 three pointers in the last five years of his career, and already has hit more three balls in the Sun's first three preseason games (3) than he hit all last season (2). I guess you can teach an old dog new tricks. I'm surprised that more fantasy pundits aren't taking note of this. Originally, people expected Hill to be a psuedo-playmaker and potentially take some of those responsibilities away from Boris Diaw. However, if the preseason is any indication, the Phoenix-version of Grant Hill is nothing more than a glorified chucker, a very old, handsomely paid chucker at that.

D'Antoni has suggested that Grant Hill might even make the all-star game this year and average 20 pts a game. Let's come off that ledge, Mike. The only way Hill makes All-Star weekend is if he's playing in that stupid two ball contest with Cedric Ceballos and some chick from the WNBA. Even with the Suns' high powered-offense, it takes some very aggressive assumptions in order to get Hill up to 20 points a game, both in production and minutes. It's hard to imagine him doing either at his advanced age of 35 (what am i talking about...I'm going to be 34 on Saturday!). I have probably the most aggressive set of projections for Hill and it tops out at about 15 pts, 1 1/2 3PA per game and about 30 minutes per contest.

One thing I do worry about with Hill is his effect on Raja Bell, Boris Diaw and Leandro Barbosa. The Suns ran with a six man rotation with Kurt Thomas and James Jones mixed in sparingly. Both those guys are gone, but now the Suns will be going seven deep with a few minutes here and there for Brian Skinner and Marcus Banks. The rest of the Hill's minutes have to come from somewhere. You figure Nash could spare a minute or two to keep him well-rested, Marion and Amare are probably going to see the same or more minutes compared to last year. The rest will have to come from Barbosa, Diaw and Raja. Raja logged 37+ minutes last year, I wouldn't expect him to be anywhere near that in 2007-08. I hope Barbosa continues to flourish, as I've drafted him in three of my four leagues so far.

- Last year, I blogged about the predictive power of the preseason, and one of the things I looked at was the league-wide preseason numbers being a leading indicator for the type of scoring, shooting, etc we could expect for the regular season. I even tested making some global adjustments to my projections based on these league-wide numbers. The global adjustment projections didn't fare as well as my individual player-based preseason adjustments, but 2006-07 was a screwy year. You had the new ball introduced right before training camp started, then saw it scrapped altogether at the start of 2007. Certainly that created some noise in the 2006-07 numbers.

So far this preseason, it looks like scoring will be up and field goal shooting will be down this season, following a trend we've seen for a few years now. Free throw shooting is up slightly relative to previous preseasons, and we generally see the regular-season FT numbers about 2%-3% higher than the preseason numbers. Maybe Dwight Howard is skewing the numbers. And look at the rate of blocks relative to preseason seasons. The initial numbers show a 26% increase in blocks for the upcoming year. Overall, the projected 2007 numbers show about a 2.2% increase in total fantasy production relative to 2006. However, it's a little too early in the preseason to make any definitive conclusions yet. It could simply be a function of the teams that have played more preseason games to date, or a function of players that have played versus sat out. But it's worth keeping an eye on. Initially, I have leveraged the preseason player data to make adjustments to the preseason, but have normalized the numbers so the weighted average totals of the adjusted numbers match the original totals. With a little more data, I will start to let the adjusted numbers follow the trends from the preseason, although the overall impact will be small.

GM_03PRE _05PRE_06PRE_07PRE _03REG_05REG_06REG REG/PRE_07REG?_07 vs O6
MIN 48 48 48 48 48 48 48 - 48 -
FGM 6.58 6.73 6.88 6.92 7.04 7.12 7.33 6.5% 7.39 0.8%
FGA 15.24 15.23 15.23 16.16 16.01 15.70 15.99 4.4% 16.91 5.8%
3PM 0.84 1.00 1.03 1.07 1.06 1.14 1.22 19.5% 1.27 4.5%
3PA 2.50 2.89 3.11 3.24 3.03 3.18 3.40 13.5% 3.67 8.0%
FTM 4.08 4.43 4.64 4.77 3.68 3.90 3.93 -12.3% 4.18 6.4%
FTA 5.59 6.06 6.32 6.46 4.90 5.23 5.23 -14.4% 5.53 5.8%
REB 8.23 8.08 8.08 8.82 8.47 8.14 8.24 1.9% 8.99 9.1%
AST 3.91 4.02 4.05 4.05 4.23 4.10 4.26 5.1% 4.26 0.0%
STL 1.85 1.78 1.71 1.76 1.58 1.42 1.45 -16.5% 1.47 1.5%
BLK 0.98 0.94 0.86 1.13 1.02 0.93 0.93 3.8% 1.17 26.3%
TO 3.56 3.48 3.74 3.68 2.84 2.74 2.91 -21.3% 2.90 -0.3%
PTS 18.07 18.89 19.43 19.68 18.82 19.28 19.81 2.7% 20.22 2.1%
2PM 5.74 5.73 5.85 5.86 5.99 5.99 6.11 4.4% 6.11 0.1%
2PA 12.74 12.35 12.11 12.93 12.98 12.52 12.59 2.4% 13.24 5.2%
FG% 43.2% 44.2% 45.2% 42.8% 44.0% 45.4% 45.8% 2.0% 43.7% -4.7%
FT% 72.9% 73.1% 73.5% 73.8% 75.2% 74.5% 75.2% 2.5% 75.6% 0.6%
3P% 33.5% 34.6% 33.0% 33.0% 34.8% 35.8% 35.8% 5.3% 34.7% -3.2%
2P% 45.0% 46.4% 48.3% 45.3% 46.1% 47.8% 48.5% 1.9% 46.2% -4.9%


Me Against The World

"Nothin to lose, it's just me against the world baby" - Tupac

There are a lot of fantasy basketball websites selling their expert projections. But how do you know you're getting something worth paying for? Especially when there may be better projections readily available for free (nudge, nudge). Well, as a man of the people, I've broken down the different options for you:

Pay for Play

ROTOWIRE.COM

Cost: $14.99 through site, $9.99 as part of yahoo draft kit (pdf only)
Depth: 355 players
Detail: Standard roto categories. No minutes, 3PA or 3P%. Shooting stats show projected cumulative totals with games played. REB, AST, etc shown in per game by tenths, but using Points League tool can get you cumulative totals to convert to more precise per game estimates. Rebounds split into offensive and defensive.
Updates: Updated periodically. Nice change log feature shows each change and rationale.
Additional Features: Customized rankings tool, 2006-07 stats in Excel format, Cheat Sheets, Depth Charts, Expert Insights and Team Previews
Past Performance: 6th of 7 in 2006-07; 5th out of 5 in 2005-06
Conservative/Aggressive: Top 210: +5.1% vs 2006 (2nd out of 11); Top 140: +5.2% vs 2006 (2nd out of 12); 12 Rookies: +8.3% vs consensus (3rd out of 11)
Bullish Vs. Consensus: Leandro Barbosa, Jeff Green, David Lee, Kenyon Martin, Kyle Lowry
Bearish Vs. Consensus: Grant Hill, Ricky Davis, Earl Watson, Joel Przybilla, Ryan Gomes
Comment: Rotowire are the folks behind the projections used for Yahoo draft kits and the numbers behind the 'O-Rank' if you play Yahoo fantasy teams. You can buy the draft kit on yahoo for $9.99 along with the stat tracker service or get a little more customization and functionality directly from their website for $14.99. Rotowire also now supplies NBA.com with contents and projections, but I think it's done by two different people as the projections are different from each other.

BASKETBALL MONSTER

Cost: $14.99
Depth: 357 players
Detail: Standard roto categories, both per game and cumulative. Minute projections are whole numbers. Game projections are in 5-game increments. Extras include splits by every stat you could possibly think of: off vs def rebounds, double-doubles, etc.
Updates: Updated periodically.
Additional Features: Lots of customization. Historical rankings from last 7 years, last 3 playoffs, last 2 preseasons, even last 2 summer leagues (deep sleeper: Toby Bailey).
Past Performance: 4th of 7 in 2006-07; 2th out of 5 in 2005-06
Conservative/Aggressive: Top 210: +9.2% vs 2006 (3rd out of 13); 12 Rookies: -8.2% vs consensus (10th out of 12)
Bullish Vs. Consensus: Bostjan Nachbar, Louis Williams, Dorell Wright, Desagana Diop, Yi Jianlian
Bearish Vs. Consensus: Kurt Thomas, Willie Green, Gerald Green, Julian Wright, Rafer Alston
Oddities: Until I pointed it out, Marco Belinelli had more 2PM than 2PA.
Comment: Basketball Monster has a bunch of useful features to help you manage your fantasy basketball squad that make it a bargain at $15 even if you don't use their projections. Customized projections is a new feature that allows you to tweak their projections for any player that you feel is out of line. Monster is particularly useful if you have non-traditional categories in your league (Dirk Nowitzki is the top player if you happen to be playing in a 26-category roto league.)

HOOPS KLYCE

Cost: $16 for initial projections, $10 more for updates
Depth: 280 players
Detail: Standard roto categories as per game averages in tenths. FG% and FT% in whole percentages. Games and minute projections included (down to half-minute increments). No 3PA or 3P% information available, however all information is available at a detailed level by request.
Updates: Updated periodically if you pay extra.
Additional Features: Draft kit is a big Word doc with rankings, player profiles, expert insights. Preseason update package includes e-mail newsletter highlighting preseason performance from players that may lead to projection changes. Additional pay-for-features available (auction leagues, additional teams, etc.)
Past Performance: Not included in my projection analysis, but he edged me out in rotosource's competition last year.
Conservative/Aggressive: Top 210: +8.1% vs 2006 (5th out of 13); 12 Rookies: +10.2% vs consensus (2nd out of 12)
Bullish Vs. Consensus: Tyrus Thomas, Kenyon Martin, Louis Williams, Ronnie Brewer, Nick Young
Bearish Vs. Consensus: Shaquille O'Neal, Smush Parker, Alonzo Mourning, Cuttino Mobley, Kyle Korver
Comment: Hoops Klyce is one of the most expensive draft resources out there, but he has a good track record.

ROTOWORLD

Cost: 'Only' $14.99
Depth: 355 players
Detail: Same detail as rotowire...standard roto categories. No minutes, 3PA or 3P%. Shooting stats show projected cumulative totals with games played. REB, AST, etc shown in per game by tenths, but using Points League tool can get you cumulative totals to convert to more precise per game estimates (not an easy task)
Updates: Updated periodically.
Additional Features: Export to PDF or printer-friendly format (says you can export to Excel but I haven't found that feature), player profiles, cheat sheets, expert analysis, customized rankings.
Past Performance: Not included in my projection analysis
Conservative/Aggressive: Top 210: +9.2% vs 2006 (2nd out of 13); 12 Rookies: -5.3% vs consensus (8th out of 12)
Bullish Vs. Consensus: Delonte West, DeShawn Stevenson, Mike James, Steve Francis, Kwame Brown
Bearish Vs. Consensus: Ben Wallace, Corey Brewer, Charlie Villanueva, Walter Herrmann, Jeff Green
Comment: Rotoworld is a great place for player updates. It's head hoops guy, Dr. A, seems to know his stuff and does well in all those fantasy 'expert' leagues. The content and functionality is very similar to Rotowire (and for the same price). I tried to get their stat projections for 2006-07 to see how they stacked up, but to no avail. They were in rotosource's analysis, but did not make the top 3.

FANTASY LOUNGE

Cost: $7.99
Depth: 372 players
Detail: Basic per game averages down to the tenths. Minutes in whole numbers. Some percentages in whole percents (varies by team). No shot attempt numbers, no 3PA or 3PP, no games played projections.
Updates: Does not appear to be updated. Most projections posted in mid-September. Greg Oden projection to average 10.6/8.5 in 25 minutes, for example.
Additional Features: Team previews, expert analysis.
Past Performance: First year of projections, I believe
Conservative/Aggressive: Top 210: +1.6% vs 2006 (13th out of 13); 12 Rookies: -24.8% vs consensus (12th out of 12)
Bullish Vs. Consensus: Shaquille O'Neal, Rajon Rondo, Raymond Felton, Walter Herrmann, Kurt Thomas
Bearish Vs. Consensus: Julian Wright, James Posey, Craig Smith, Jeff Green, Boris Diaw
Comment: Fantasy Lounge's draft kit is not the easiest to use. In order to get the projections, you have to download each team's preview individually and cut and paste from a pdf file. I tried to get their projections posted as one lump Excel file, but only got the response: 'That's a good idea for next year'. Plus for some reason their forums cause my browser to stall out and crash every 2-or-3 clicks. Just getting the full set of projections was a 2-hour process. Lounge does a good job of projecting out minutes for the full 12+ man roster, however for many teams they are careful to make sure that the minutes only add up to 240 total. This would only make sense if each guy was expected to play 82 minutes. For starters, I think their minute projections are okay, but I think they forced the minutes for bench players down to fit this 240-minute constraint. That might be one reason why their rookie projections are far more conservative than the consensus (although they could be the most accurate).

FANTASY HOOPS REALM

Cost: $14.95
Depth: 267 players
Detail: Detailed per game averages down to the tenths. Minutes and games played projected. Extras include offensive and defensive rebounds and personal fouls.
Updates: Does not appear to be updated. Sean May is still logging minutes for Charlotte
Additional Features: Player profiles, cheat sheets, expert analysis
Past Performance: Not included in past analysis, not sure if this is their first year or not.
Conservative/Aggressive: Top 210: +8.2% vs 2006 (4th out of 13); 12 Rookies: -10.8% vs consensus (11th out of 12)
Bullish Vs. Consensus: Shaquille O'Neal, Rajon Rondo, Raymond Felton, Walter Herrmann, Kurt Thomas
Bearish Vs. Consensus: Julian Wright, James Posey, Craig Smith, Jeff Green, Boris Diaw
Comment: Stumbled across the Realm from a google ad (see, they really do work...feel free to click on any (or all) of mine to the right of this column). Looks like a pretty solid effort, good level of detail, lots of analysis. I'd like to see more updates and a deeper player pool.

The Freebies

CBS SPORTS

Depth: 483 players
Detail: Standard roto categories. No minutes, no 3PA but can be extapolated from 3P%. Shooting stats show projected cumulative totals with games played. REB, AST, etc shown in per game by hundredths.
Updates: Updated periodically, but sparingly. I noticed Leandro Barbosa was moved up and they made the necessary changes for season-ending injuries to Oden, Etan Thomas and Sean May. Other than that, projections are fixed.
Additional Features: None
Past Performance: 7th of 7 in 2006-07; 4th out of 5 in 2005-06. Only site with the distinction of performing worse than simply using the prior year's stats two years in a row!
Conservative/Aggressive: Top 210: +1.8% vs 2006 (12th out of 13); 12 Rookies: +6.0% vs consensus (5th out of 12)
Bullish Vs. Consensus: Yi Jianlian, Danny Granger, Charlie Bell, Luther Head, Kyle Korver
Bearish Vs. Consensus: Luol Deng, Dorell Wright, Ronnie Brewer, Jason Maxiell, Nene Hilario
Oddities: Antonio Daniels projected at 87 games, Jason Terry projected at 86 games. 483 players deep, but no Luis Scola
Comment: You get these projections free when you sign up for a team, and if history is any indication, you get what you pay for.

ESPN

Depth: 293 players
Detail: Standard roto categories on a per game basis in tenths. No minutes, no games played, no 3PA or 3P%.
Updates: ESPN got a late start on their fantasy offerings, so their projections were just uploaded recently. Not sure if they will be updated (I'm guessing not).
Additional Features:Player profiles, team previews, mock drafts, Matthew Berry's 'expert' Draft-Day Manifesto.
Past Performance: 6th of 7 in 2006-07 (as TMR); 3rd out of 5 in 2005-06 (as TMR on NBA.com).
Conservative/Aggressive: Top 210: +4.0% vs 2006 (8th out of 13); 12 Rookies: +3.8% vs consensus (6th out of 12)
Bullish Vs. Consensus: Kyle Lowry, Walter Herrmann, Allen Iverson, Josh Childress, Ronnie Brewer
Bearish Vs. Consensus: Kenyon Martin, Dorell Wright, Jarrett Jack, Bobby Simmons, Luke Walton
ESPN bought out Talented Mr Roto so now Berry's running all of their fantasy stuff. The good news is you don't have to pay for their projections any more, and as I've shown multiple times, they are not worth much. I'm disappointed we have yet to see Berry's 'signature' Love-Hate list for the upcoming season. I was looking forward to seeing him push Speedy Claxton and Desmond Mason and defend those as good picks.

82GAMES.COM

Depth: 150 players right now, looks like they may have 200 soon
Detail: Basic per game averages down to the tenths. No minutes or games played. Percentages in tenths of a percent, but no shot attempts and no 3P%. No turnovers is a big turn-off to those who play 9-cat leagues.
Updates: 82games just launched its fantasy section. Not sure if they'll be making updates.
Additional Features:Player profiles, expert analysis
Past Performance: New Projections
Conservative/Aggressive: Top 140: +11.3% vs 2006 (1st out of 13); 12 Rookies: +8.5% vs consensus (4th out of 12)
Bullish Vs. Consensus: Dorell Wright, Jason Maxiell, Joel Przybilla, Jason Kapono, Ronnie Brewer
Bearish Vs. Consensus: Cuttino Mobley, Chris Kaman, Al Thornton, Antawn Jamison, Luke Ridnour
Comment: 82games.com is a great site for basketball sabrematicians. Their fantasy section is simply one guy's take on the upcoming fantasy year. I'd like to see them explore whether they can leverage some of the other analysis from the site into their fantasy projections.

NBA.COM

Depth: 349 players
Detail: Basic per game averages down to the tenths. Percentages in tenths of a percent, but no shot attempts and no 3P%. No turnovers. Games Played and Minutes (in whole minutes) projected.
Updates: None
Additional Features:Player profiles, expert analysis, cheat sheets, etc.
Past Performance: 6th out of 7 in 2006-07; 3rd out of 5 in 2005-06; both supplied by Talented Mr Roto
Conservative/Aggressive: Top 210: +3.7% vs 2006 (9th out of 13); 12 Rookies: +24.0% vs consensus (1st out of 12)
Bullish Vs. Consensus: Corey Brewer, Marco Belinelli, Andrew Bynum, Al Jefferson, Acie Law
Bearish Vs. Consensus: Ronnie Brewer, Louis Williams, Joel Przybilla, Morris Peterson, Tarence Kinsey
Comment: NBA.com has migrated to Rotowire after using TMR the last few years. Jon Loomer completely revamped their fantasy offerings last year to mixed reviews and has made additional changes this year. I don't know if the projections are Loomer's or straight from Rotowire. It'd be nice to see a little more detail in their projections. But one thing is for sure...they love the rookies!

FANTASY SPORTS CENTRAL

Depth: 407 players
Detail: Full detailed cumulative stats with minutes and game played, so you can get detailed per game averages. Minute projections are in full minutes.
Updates: I think so, as it says rookies will be added as more preseason games are played.
Additional Features:Some customization if you sign-up for a free membership. Links to other sites.
Past Performance: 3rd out of 7 in 2006-07
Conservative/Aggressive: Top 210: +3.2% vs 2006 (11th out of 13); 12 Rookies: Excluded Rookies
Bullish Vs. Consensus: Rafer Alston, Chris Wilcox, Ben Wallace, Allen Iverson, Matt Carroll
Bearish Vs. Consensus: Monta Ellis, Jason Maxiell, Andrea Bargnani, Al Jefferson, Kevin Martin
Oddities: 407 players projected, but no Adam Morrison
Comment: Fantasy Sports Central provides the level of detail you should be looking for in your projections. However, excluding rookies leaves a big gap if you have a draft coming up soon.

ROTO FREAK

Depth: 276 players
Detail: Standard roto categories on a per game basis in tenths. No minutes, no games played, no 3PA or 3P%. Shooting percentages down to half-percents.
Updates: Not sure
Additional Features:Expert analysis, cheat sheets
Past Performance: 2nd out of 7 in 2006-07; 3rd in Rotosource competition last year
Conservative/Aggressive: Top 210: +3.5% vs 2006 (10th out of 13); 12 Rookies: -4.9% vs consensus (7th out of 12)
Bullish Vs. Consensus: Daniel Gibson, Gerald Green, Kwame Brown, Kevin Garnett, Luke Ridnour
Bearish Vs. Consensus: Yi Jianlian, Kyle Lowry, Corey Brewer, Brad Miller, Rasho Nesterovic
Comment: RotoFreak performed well last year. However, I hear someone else is doing their projection's this season.

COLTON INDEX / WEGOBLOGGER31

Depth: 341 players
Detail: Full detailed per game stats with minutes and game played. Rebounds split into offensive and defensive.
Updates: Weekly
Additional Features:Expert analysis on the blog
Past Performance: 1st out of 7 in 2006-07; 1st out of 5 in 2005-06; 2nd in Rotosource competition last year
Conservative/Aggressive: Top 210: +4.5% vs 2006 (7th out of 13); 12 Rookies: -5.8% vs consensus (9th out of 12)
Bullish Vs. Consensus: Ronnie Brewer, Louis Williams, Trevor Ariza, JJ Redick, Tyrus Thomas
Bearish Vs. Consensus: Acie Law, Corey Brewer, Donyell Marshall, Luther Head, Larry Hughes
Comment: Continually looking to improve accuracy of projections. Incorporating a bunch of analysis performed in the offseason. Trying to get better at projecting games played. Only time will tell if all that analysis was worth anything.

One thing I want to do is create a consensus set of projections that would combine all of the projections, throwing out the highest and the lowest and averaging the rest. I think of this as the 'market' projection for each player. I'm interested to see how a) the consensus performs, because I speculate that it might actually end up being the best projection set when it's all said and done; and b) how my projections stack up against the combined wisdom of the so-called fantasy experts. I think it would be quite a challenge. Anybody can beat up on TMR year in and year out. Stay tuned and I'll post the players with the most differences in the near future.

0 comments:

 
Wegoblogger #31 © 2011 | Designed by Bingo Cash, in collaboration with Modern Warfare 3, VPS Hosting and Compare Web Hosting