| 0 |
Subject: Value Players & Roster $ Distribution
Posted by: OSU Rules
- Leader [15372315] Fri, Dec 14, 2001, 13:58
This post comes from noticing a few things and getting curious about where I should be spending my precious roster $'s. I think the muted gains, changing strong schedules for studs by position and Guru's earlier comments on forwards vs. guards all make roster $ distribution by position critical.
I think right now it is very tempting to load up on some expensive forwards, but that means you need some good values at the other positions.
For fun I used the sortables to create a list of players ranked by SWP/g/$. I filtered out anyone below 20 SWP/g YTD and that had not played a game in the last 7 days. The top 25 broke out by position as follows:
8 Guards
10 Forwards
7 Centers
It surprised me how balance this was. I was also surprised by the number of Centers. Interestingly, if I changed it to the last 15 days a net of 2 Centers dropped off and 2 forwards moved up. So the Centers are probably less consistent.
What was more interesting was then projecting for the next 21 days. The top 25 in projected points/$M, but at least averaging more than 20 SWP/g in the last 15 days broke out as follows:
9 G
12 F
4 C
6 of the top 10 were forwards.
For total points, the projection has:
11 G
10 F
4 C
There are 3 players that make both lists Gasol, J. Williams, and Lamond Murray. Tinsley just misses ranking 26th in total points.
My take away from this is to stay cheap at Center, there are 3-4 reliable alternatives for 2 roster spots. Go cheap at Forward with one stud, since there are 5-6 reliable value choices for the other 3 forward positions. Finally, go expensive with Guards, probably 2 studs, one medium (~$5M) and Tinsley.
In the past, I have always thought this games was about the forwards, but I am becoming convinced that this is the year of the Guard and managing that position will be most critical. Finally, with muted gains I am also starting to wonder if Shaq is a luxury that can no longer be afforded.
My $0.02 before heading into the Holiday Season.
|
| 1 | Great One
ID: 491025139
Fri, Dec 14, 2001, 14:25
|
The obvious reason is that you can get great bang for the buck at F and its just not there at G... other than Tinsley. you can get 30-40 from guys like Gasol or 5-10 points more per game with guys like Garnet for 3x the price.
I think I am gonna have to go with Anthony and Q.Richardson(1/1 tinsley>quentin) after the break.. so I can afford Kobe, Tmac, Brand, Walker, and Shaq post christmas.
|
| 2 | Hillbilly Delight
ID: 221039714
Fri, Dec 14, 2001, 15:14
|
Yes, but if you look at the teams with the loaded schedule after Christmas (Nets, Clippers, Bulls, Celtics, Bucks, and Spurs), you can see that your top guns are mostly forwards:
Brand, Odom, Pierce, Walker, Duncan.
Of these teams only the Nets and Bucks have guards. From the way they have been playing, I will probably end up with Duncan, Brand (or Odom) and Walker.
There are some steals at the guard position too. What guards would a person get in order to gain the points Walker, Brand, and Duncan can get?
|
| 3 | Tomcraw
ID: 289503117
Fri, Dec 14, 2001, 15:15
|
Jason Kidd.
|
| 4 | culdeus
Donor
ID: 51014269
Fri, Dec 14, 2001, 15:18
|
HD, he is talking about SWPTS/GM/$$ a different criteria than total points. His data is not schedule related as far as I can tell. I think the F's you listed would have a very hard time meeting his criteria for top 25.
|
| 5 | Hillbilly Delight
ID: 221039714
Fri, Dec 14, 2001, 15:18
|
Yes, Kidd, but that is only 1 spot. If I go with cheap forwards, then Kidd only fills one of my guard spots. My point is that it may be better to load up on Walker, Brand (or Odom), Duncan, and Nailon in the forward spots. Then, hold these for weeks.
|
| 6 | Blooki
ID: 359321514
Fri, Dec 14, 2001, 15:20
|
Davis, McGrady, Iverson...
|
| 7 | Hillbilly Delight
ID: 221039714
Fri, Dec 14, 2001, 15:26
|
Blooki, none of those 3 players actually play for the teams I listed above in the teams with the greatest number of games. Thanks though.
|
| 8 | Blooki
ID: 359321514
Fri, Dec 14, 2001, 15:29
|
Kobe, Cassell, Allen, Carter, Stockton... granted some of those won't produce as much as Brand, Duncan, Garnett, Walker, Pierce, etc. per game, they will produce similarly per game per $mil.
|
| 9 | Blooki
ID: 359321514
Fri, Dec 14, 2001, 15:32
|
Stackhouse.
|
| 10 | Blooki
ID: 359321514
Fri, Dec 14, 2001, 15:33
|
Brandon, Nash... (and I still believe in Stoudamire ;þ)
|
| 11 | Blooki
ID: 359321514
Fri, Dec 14, 2001, 15:33
|
Miller.
|
| 12 | Hillbilly Delight
ID: 221039714
Fri, Dec 14, 2001, 15:36
|
Blooki, do you believe a team with those guards and cheap forwards will beat a team with Duncan, Brand, Walker, and cheap guards w/ Kidd? This is the question I am asking.
|
| 14 | Blooki
ID: 359321514
Fri, Dec 14, 2001, 15:50
|
It totally depends on what time period you're asking about too. I think it's too early to load up on Fs. One or two is OK, but putting almost 30mil into Fs at this point in time when guys like McGrady and Davis still have strong schedules.
For one thing, I don't think Davis should've been sold by anyone yet... IMO, he's the best stud to own out there. Unfortunately, I only have him on one team because getting him on other teams would be a lateral move.
Come the new year, I'm definitely going after Odom and Brand. Duncan if I can. Walker, Pierce, and Garnett only if they show me more. Personally, I don't think you can plan that far ahead yet. I think planning about two weeks in advance is good, but anything beyond is probably going to be changed anyways. So arguing over this is pretty futile. What happens if Duncan goes down? What happens if Stoudamire keeps playing like he has? What happens if Shaq keeps playing like he has? You really going to try to fit Shaq, Duncan, Walker, and Brand under the cap? Good luck...
|
| 15 | Pilewort
Donor
ID: 369101914
Fri, Dec 14, 2001, 15:54
|
OSU Rules: Interesting. Maybe I need to be more into cheap guards.
I wish there were an idiot filter in the sortables. That way I would have avoided Derrik Coleman and the human headache (J. Williams). First the guy's on drugs, then when he needs to be, he's not.
I haven't a clue how to play this game, but if anybody's got an optimizer routine that employs a 2-week schedule and a 15-day SNP/G ave., I'm all ears.
|
| 16 | Hillbilly Delight
ID: 221039714
Fri, Dec 14, 2001, 15:54
|
Blooki, sorry. I am thinking about the time period directly after Christmas. I too have Davis and TMac, and there is no way I am dropping these guys before Christmas.
A TMac --> Kidd trade is tempting, but I think if TMac continues to dominate I will wait until January 9th before I drop TMac.
What are you planning on doing with Davis?
|
| 17 | Blooki
ID: 359321514
Fri, Dec 14, 2001, 15:56
|
Davis can be held until the 29th as far as I'm concerned. I will only move him before then if I really need the money he's tying up.
|
| 18 | Special K @College
ID: 291161414
Fri, Dec 14, 2001, 15:57
|
Blooki, I believe in Stoudamire too. :)
|
| 19 | Tomcraw
ID: 289503117
Fri, Dec 14, 2001, 16:22
|
I am probably going to stick with 2 stud guards and 2 stud forwards.
Kidd + somebody (maybe Carter)
Duncan + Brand
I have Kidd and Duncan as of today, and will get the other too later. Still holding Davis and TMac.
It is just too hard to come up with 3 cheapie guards to justify picking up a Celtic to go along at F.
|
| 20 | DR Stars
ID: 162592010
Fri, Dec 14, 2001, 16:23
|
Pilewort there is one, I can email it to you (excel file using the sartibles), but if you don't hurry (I'm getting of from work soon) it'll have to be on monday, post your email in here and I'll send it.
|
| 21 | Guru
ID: 330592710
Fri, Dec 14, 2001, 16:29
|
Just for kicks, I used the excel solver to see what the optimum $60 million roster would look like for the month of January. To project points, I used today's prices, and TSNP/G for the past 15 days. The object was to find the static roster, costing $60m or less, that produces the maximum number of points from 1/1-1/31.
Here's what the solver came up with (number of games during the month in brackets):
Guards: Tinsley[14], Q. Richardson[18], Person[15], Griffin [15]
Forwards: Duncan[16], Brand[18], Odom[18], Murray[15]
Center: Shaq[16], Mutombo[15]
Interesting. Four cheap guards ($8.7m total), four expensive forwards ($31m), two expensive centers ($20.2). I wasn't surprised to see Brand and Odom, given their schedule advantage (2 more games than any other team). This roster projects out at 5876 points.
However, you only sacrifice 2 points (5874) by swapping out Brand and Person for Vince Carter and Kirilenko. So don't over-stress the expensive forward theme here. In fact, if Kirilenko's average were just 0.2 higher, then Brand wouldn't have even been on the optimal roster. Ditto if Brand's average were slightly less (and given that he'll have 6 sets of back-to-back games in January, a slight drop-off doesn't seem out of the question.
Another caveat. Sometimes the solver doesn't quite find the maximum roster. But I'll bet this is close, given the assumptions I used.
Of course, you don't need to hold a static roster for the month. And you need to find an efficient way to get from where you are to where you want to be. Swapping from a guard-dominated roster to a forward/center dominated roster isn't particularly easy.
But this does seem to confirm the hypothesis above - that once the schedule goes light for the top guards, it may not be such a bad idea to move your bucks into forwards, in spite of the cheap alternatives at that position.
Certainly food for thought.
|
| 22 | Species
Donor
ID: 304521510
Fri, Dec 14, 2001, 16:34
|
Outstanding post OSU Rules.
I had pondered the same question as you. As everyone else seemed to be loading up at F with 2 studs, I found that the cheap forwards I already had (Nailon, Kirilenko and Gasol) were producing so strongly for their prices that it would behoove me to go big at Guard with McGrady, Baron, Brandon and (this is a stretch) Stockton (might dump him for Tinsley after tonight). Who knows what is truly the best move, but I like your thought process and came to roughly the same conclusion....I just skipped the analysis ;-).
|
| 23 | Blooki
ID: 359321514
Fri, Dec 14, 2001, 16:40
|
Shaq/Mutombo, eh? Neat how one of my teams already has the two-headed monster in place (I know it's a little early, but I had the cash). Unfortunately, this team is looming around a WWR of 700. ::sigh::
|
| 24 | Guru
ID: 330592710
Fri, Dec 14, 2001, 16:47
|
A team ahead of its time...
|
| 25 | TaRhEElKiD
ID: 42109719
Fri, Dec 14, 2001, 16:51
|
Here in a few week I will have these studs:
Kidd
Brand
Shaq
Duncan *Have now*
Rest of the team:
Tinsley
Gasol
Kirilenko
TMac2
Griffin
Parker or Q or Maggette (?)
|
| 26 | Hillbilly Delight
ID: 221039714
Fri, Dec 14, 2001, 17:00
|
TaRhEElKiD, my projected roster is looking very similar. I too am deciding b/w Q-Richardson and another guard, but I am not sure yet. I will probably go with Walker instead of using the $$ for O'Neal. Probably a bad move, but I can't see spending $12 million for O'Neal when I have Duncan at $11 million.
About Kirilenko. I am not familiar with the Jazz this season, but if Russell comes back (and I hear it will be soon) will he affect Kirilenko's playing time?
Also, Guru, where can I get that optimal program?
|
| 27 | OSU Rules
Leader
ID: 15372315
Fri, Dec 14, 2001, 17:04
|
One way to look at the guard vs. forward debate is the following:
Garnett, Duncan, Webber, Walker are maybe 2-3 pts/g better than the comparable top guards (Bryant, Kidd, Iverson, Payton, McGrady, Davis). However, if you assume you need 2-3 cheapies, the cheapie forwards will be 4-5 pts better. Thus, if you are going to have 4 studs and 4 cheapies you will do better with 3 stud guards and 3 cheapie forwards than vice versa.
I did not talk about schedule directly, because I was trying to determine how to divide up my roster. In my mind, there will be a pool of players in each category that I can then optimize for schedule. What is difficult is going from 3 stud guards to 3 stud forwards, because that takes 4-5 trades rather than 3.
If a Stud= >40 SWP/g
Over the last 30 days, 9 G's met the criteria and only 5 forwards. Thus, I should always be able to find 2-3 guards out of the 9 with good schedules and I can use my trades to rotate through them efficiently. The forwards are thin, so I may be forced to move money around, because there are no stud forwards with good schedules or I need money to upgrade to one, because they all have good schedules. That seems inefficient to me.
It is important to think about the cost
Another interesting piece of data:
Top 25 in SWP/g for <$4M
6 G (including Tinsley, by far #1)
14 F (including 7 of top 9)
5 C
Given muted gains, I think the salary slots on my roster are going to effectively stay constant. Looking at these numbers, I think I should structure my roster to have 2 stud guards plus Tinsley and then a cheapie, 1 stud forward and 3 value forwards, and 2 value Centers. Over time I might be able to move up the forwards, but I don't see having multiple stud forwards.
|
| 28 | Guru
ID: 330592710
Fri, Dec 14, 2001, 18:21
|
There is a link to a downloadable version of the basic optimizer in post [9] of this thread. I tried to make it rather self explanatory, but there are a number of other comments inm that thread, too.
|
| 29 | rage_22@work
ID: 57932917
Wed, Jan 30, 2002, 17:26
|
BUTT, has this changed over the past 6 weeks?
|
| Rate this thread: | If you wish, you may rate this thread on scale of 1-5. Ratings should indicate how valuable or interesting you believe this thread would be to other users of this forum. A '5' means that this thread is a 'must read'. A '1' means that this is a complete waste of time. If you have previously rated this thread, rating it again will delete your previous rating. If you do not want to rate this thread, but want to see how others have rated it, then click the button without entering a rating, or else click here. |
|
|
Post a reply to this message: (But first, how about checking out this sponsor?)
|
|