SubscribeSince the start of fantasy baseball, the constant challenge that participants have faced is trying to determine players’ values. What do you pay for a guy who can get you 30 SB? Is he worth more or less than someone who hits 30 HR? Would you trade him for someone with a 3.65 ERA?
Really, it’s a problem that boils down to this: How do you measuring a player’s contribution, when the contributions are based on 8 or 10 different scales (i.e. the 4×4 or 5×5 stat categories)? The solution is to convert all of the categories to use a common scale.
One way to do this is to use Standings Gain Points. Standings Gain Points are a system developed by Alex Patton and popularized by the recent re-release of Art McGee’s How To Value Players For Rotisserie Baseball. Derek Carty praises SGP as “the greatest, most logical way of valuing players I’ve ever heard of.”
How do SGPs work?
The basic idea behind SGP is to look at the final standings from your league in previous years and determine how much of each stat it takes to pick up an extra point in the standings. How many homeruns would it take to move up one spot in the homerun standings? How many strikeouts separate each team?
If you notice that there are 10 HR separating each team, then you know that, on average, it will take an additional 10 HR to move up one point in the standings. So 10 HR are equal to 1 SGP. Do the same thing for all of the other stats, and all of a sudden you have reduced a player’s contributions from across four or five categories down to one, simple number. Sort players by that magic number and you will have rankings customized for your draft. Convert that number into a dollar value and you are ready for an auction.
To be honest, I think Standings Gain Points is overall a pretty sound methodology. However, I do think it has some flaws that keep it from being “the greatest, most logical way of valuing players” ever.
This week I’ll be taking a look at some of the problems I see with SGP, and also explain why I decided to take a different approach when building my own valuation system, the Price Guide.
Related posts:
As with any valuation system, SGP has its issues i.e. utter dependence on projection accuracy after accepting that the SGP denominators are a constant.
I set-up a SGP model using my existing keeper league standings. Fortunately, AB, IP, ER, etc are there.
My denominators are very different from thos in McGee’s book esp in the AL.
One of the biggest flaws with SGP is that it’s great to value what it takes to move from 12th to 11th in a category, for example, but what about accumulating the stats to even reach the typical last place total to begin with?
SGP ends up undervaluing stolen bases, and valuing them similarly to HRs, since the denominators are close to each other. Yet, it doesn’t take into account the fact that the supply of SBs is a lot less than HRs. And economics teaches us that less supply will increase demand (or price), so SBs should not be worth the same as HRs.
Just to clarify, the key part of that statement is “that I’ve ever heard of.” SGPs certainly have their flaws, as do all systems – I’ve never denied that – but my point was that I haven’t seen a system that I like better, You can make some adjustments to help fix some of the flaws of SGPs.
@Mike: SGPs are assuming that a smaller pool of SB will make the categories tighter, and thus make the denominator smaller. McGee’s denominators give about a 20% premium for SB over HR in an AL/NL only league. I get about the same difference when using standard scores, and for a mixed league SBs end up worth about 35% more.
Would you value them even higher?
@Derek: I’m just teasing, but I did like your use of superlatives. :-)
It depends on the position using my valuation system (well Todd Zola’s), but in the OF, BJ Upton’s projected 35 SBs are worth $14.47 and Carlos Lee’s projected 35 HRs are worth $8.13. So SBs here are worth 78% more. At 1B/3B, SBs are worth even more since replacement level is lower. Interestingly, a HR from a 2B/SS is worth more than an SB! This is the opposite from 1B/3B in that it’s easy to find SBs from an MI on free agency. Last, SBs at catcher are worth the most since it gives you a huge advantage when no one else could get those from the positon.
@Mike: I’ll admit that I had never thought of valuing categories differently for different positions. It sounds like a very good idea, though.
I guess I’ve always considered the ease of acquiring SBs from free agent MI would be balanced by an equal ease of acquiring power stats from free agent CI… Definitely some food for thought.
Yup, but I can’t take credit for it, it’s all Todd Zola’s system that I use. I’m not smart enough to create my own :-)
The problem with evaluating stats differently for each position is that you end up with predictions like Chone Figgins is a god. This type of system would say that his 40 steals as a 3B are better than 60 HR from 3B.
40 steals from 3rd base is the same as 40 steals from 2nd base or 40 steals from a catcher when it comes to winning your league.
In fact, often times getting 40 steals from 3B and foregoing say 20 HRs from an average 3B puts you behind the 8-ball when it comes to power stats, since you need to make up for it elsewhere.
In the end, if a steal is worth 1.2 HR, then someone with a 100 R, .300 AVG, 80 RBI, 10 HR, 0 SB line is worth exactly the same as someone with a 100 R, .300 AVG, 80 RBI, 0 HR, 12 SB line, whether that player plays 3B or 2B or C.
At least, I’ve never seen a convincing argument otherwise.
Well no Molson, Figgins is only worth about $12 in my latest values and 95th overall. That certainly isn’t god-like!
Anyway, it again comes down to replacement level. We know that 1st basemen on the whole provide much better raw stats than catchers, so we adjust accordingly. But we could take that a step further and say exactly which categories are easier and tougher to find from a free agent at each position.
Saying that 40 steals from 3B is the same as 40 steals from 2nd base is the same as valuing a catcher and 1st baseman the same who put up the exact same raw stats. Basically, what Todd’s valuation system does is quantify replacement level everywhere rather than just one number for each position.
If you think of a points league, it’s easy to value players. You simply add up the total points and subtract the next best non-positive valued player’s point total from everyone at the position. This would be equivalent to a valuation system that doesn’t adjust every category by position, but just uses one overall number.
And damn, this stuff is so hard to explain! Whether you agree it’s correct or not, I just hope I’m explaining the system well enough so you guys understand what I’m talking about.
I understand the idea behind the system, I’m just not convinced (yet?) that it’s correct or useful.
The fact that Figgins is only $12 reassures me a little.
Does the positional stat adjustment base things on z-scores, or does it do a SGP-type analysis?
Maybe I’ll crunch some numbers on my spreadsheet (I don’t have as pretty a program as Mays for generating dollar values, sad to say) and see what I can get.
Related to SGP, does anyone have a resource for providing average final standings, as well as standard deviations and ranges, for fantasy leagues in various configurations? I’ve looked for this before but I’ve not been able to find something that says “the average HR score in a 12-team Yahoo! style mixed league is 200, the SD among HR scores is 20, the average range is 75″ or whatever. The limited data that I have is my fantasy leagues, and that’s limited to about 20 leagues or so over the last 5-6 years, which isn’t very useful.
The adjustments are not based on SGP or Z-Scores actually. They simply subtract replacement level for the counting stats and then the ratios are converted into counting stats with the same adjustment made.
Ah, I see. This is exactly the same thing that the z-score system does, more or less.
It’s not a different value for counting stats for each type of player, so a SB is worth the same from a SS or a C, but it adjusts the stats such that it assumes you have a team full of replacement level players and only gives points for marginal stats.
I assume for this you have to take some sort of average level for replacement players. Otherwise, what if Figgins was the first non-drafted 3B? Or half the replacement OFs are speed guys, and half are power hitters?
Think of it this way…
The price guide determines the value of each stat by looking at what the average is for each stat among drafted players and how the stats are distributed among them. In the end, the stats of different players add up to give final standings, with an average and a distribution of the stats.
The SGP system takes known final standings and a known distribution to determine the relative value of a stat. So the way that the two systems value stats is the same.
The price guide then looks at each player and figures out how many points their stats are worth, and then subtracts out a baseline in each position. That baseline is simply who it determines to be the most valuable replacement player.
The SGP system does this in reverse order. It subtracts out what it determines to be replacment level player at each position, and then calculates the points that each player’s stats generate.
Since I assume that your baseline replacement player has to be an average of some sort, these systems do exactly the same thing.
The difference is in how it determines what each stat is worth relative to each other, which is really what you’re trying to get at with valuation systems in general. One uses standard deviations to estimate it, the other uses known variance in final standings. I’d guess that the ratio of SB value to HR value (or pick any two stats you want) is pretty close between your SGP system and the z-score system on this site if the z-score system is given the same league configuration settings as your SGP system.