valuation of players for salary cap style of games

[djbob]

Hi,
I am playing salary cap type of games. i was off for a few years and i seem to be lacking a good way to analyze player values in relation the their salaries. I made a general research on google and this site came up as the one where experts could answer this type of questions.

i used back in the years to be able to get a faire value by using what cdm used to call PRM points. Calculating the PRM points per dollar spent was a very effective way to estimate a players value.

For any reason, the formulas i was usign then dont seem to be working anymore. the steals are getting way too many points and therefore guys with steals all come up with higher values. that is not untrue, but you dont want to end up with 25 points in steals and 1 in home runs and rbis.

Did someone here found a good way to estimate players values based on their stats? Somehow the formulas exist. they must use a very similar formulat to actually calculate salaries. many sites offering similar games come up with extremely similar salaries for each players, so there must be a logic behind the madness.

any info very appreciated

thanks in advance

[Todd Zola]

Hi,
I am playing salary cap type of games. i was off for a few years and i seem to be lacking a good way to analyze player values in relation the their salaries. I made a general research on google and this site came up as the one where experts could answer this type of questions.

i used back in the years to be able to get a faire value by using what cdm used to call PRM points. Calculating the PRM points per dollar spent was a very effective way to estimate a players value.

For any reason, the formulas i was usign then dont seem to be working anymore. the steals are getting way too many points and therefore guys with steals all come up with higher values. that is not untrue, but you dont want to end up with 25 points in steals and 1 in home runs and rbis.

Did someone here found a good way to estimate players values based on their stats? Somehow the formulas exist. they must use a very similar formulat to actually calculate salaries. many sites offering similar games come up with extremely similar salaries for each players, so there must be a logic behind the madness.

any info very appreciated

thanks in advance

This is something I have dabbled with but have not played salary cap games for a couple of years. Long time followers of the site may chuckle if they read this, since I am pretty well known for being an opponent of SGP style of valuation, but in this scenario, it is a better means of ranking players.

I am in the midst of a big subscription content crunch, but would like to spend some time playing with this in a coupe of weeks. Do you have a particular salary cap game in mind where the salaries are publicly available? And, do you have a rough idea of the standings in a typical league? I used to have access to the CDM data but at least as of now, no one has picked up that exact game, though there are rumblings someone may. If I can get some data, I am sure I can come up with something in early March.

[djbob]

first thanks for the very quick response.

here are a few things that might help you

A number of sites are offering similar games, but so that we dont get in all directions at the same time, the one i am most likely to play is fantrax.com the game is a carbon copy of former cdm game in all aspects. to say they are targeting the orphan customer base of cdm/fanball would be an understatement. the salaries have already been posted on the site. if you want them in excel format i can provide that too

structure is roto 5x5 with leagues of 25 players and some overall prizes. i also havent played in a few years, but back in the days, you had to aim for 215-225 to have any chance at winning your league, so it is not like you can punt some categories away

the math component in these salary cap games is quite very high. if you dont get the math part you cant compete. superstars can be unplayable if priced too high, and average players with 15 hr and 60 rbis can be perfectly fine if priced low enough. it all comes down to trying to draw a line between projected stats and actual values. doing this lets you see who are the bargains and who are the overpriced. this might be easy to calculate in a points game, but not as much in regular roto. the value of straight stats like hr, rbi, r, w, s is not so complex. the value of era, whip and avg is much more difficult as there are negative values involved.

what i am looking for is a model that would translate stats into a numerical value. comparing this numerical value to the salaries would then be a formality

you can probably break this type of game because of the very high math and analysis component



thank you in advance.

[Todd Zola]

There are negative values involved with respect to the counting stats as well.

Do you have representative standings from a CDM type league? Actually, I know enough people that I can get those.

The basic idea is you figure out how many of each stat it takes to finish at "x" place in the standings. You then assign value proportionally to the player contribution towards that goal.

As you suggest, you need to convert the ratio stats (BA, ERA, WHIP) to counting stats, but there are a few ways this can be done.

But, you also need to be careful and not just take players giving the best bang for the buck, but you need balance as well.

Most people work backwards and find the cheap guys they feel will be worth more than their salary and then fill in with the high priced guys, watching stuff like 6 games versus 7 games, a series at Coors and at Phoenix as opposed to someone playing a series in PETCO and AT&T.

The other thing I have been told by people very successful in this format, and I know 3 guys who have won the CDM challenge over the past 10 years is you cannot skimp on starting pitching -- you need to pay for the studs, regardless and find value elsewhere.