help with VORP

[mkultra]

Personally, I like the idea of "relative value", but I think obsessing over VORP as a calculated value in determining picks is WAY overthinking it. Personally, I've moved away from looking at players in a single line and instead look at it by position. Once you do that, you can see tiers clearly emerge.

You don't need a calculator to see that, although Pujols will put up better stats overall, Hanley provides better value as a #1 pick. If, on the other hand, you've got a late pick and the top 3 SS guys are already gone by the time it gets to you (though it's debatable whether Rollins is a first rounder), you can let a few more SS picks go by before you need to snag one. If you're doing 5x5, you also gain the luxury of picking a player at that position based on stat needs (speed vs. power).

The Cafe actually has an excellent set of position-based-tiers in the Rankings area to show you what I mean. I don't agree with all of their results, but it's an excellent starting point.

[RynoDyno]

The thing that confuses me with tiered rankings...say for example you go by the 1st tier of SP's listed in that article. Going by stdev methods, I get the following Z-scores:

Name	   Pos       Wpts	 SVpts Kpts   ERApts	WHIPpts	Z-Score
Santana,Johan	SP	3.62	0.00	4.21	2.86	2.56	13.41
Sabathia,C.C.	SP	3.85	0.00	4.21	1.84	2.47	12.36
Lincecum,Tim	SP	3.17	0.00	4.47	2.42	1.60	11.66
Halladay,Roy	SP	3.40	0.00	3.35	1.46	2.56	10.77
Hamels,Cole	SP	3.17	0.00	3.68	1.12	2.59	10.56
Peavy,Jake	SP	2.94	0.00	3.78	1.90	1.68	10.31
Webb,Brandon	SP	3.62	0.00	3.48	1.46	0.89	9.45
Haren,Dan	SP	3.17	0.00	3.80	0.66	1.66	9.29

A score of 13.406 for Johan and the bottom of the tier has Haren at 9.29. How does that work into "tier 1"? That seems like a huge drop considering the next 8 pitchers only have a difference of 1.533 between their z-scores. Also, without comparing all players (hitters and batters) across the board to each other, how would you start to determine what round you should be drafting these guys in?

I'm really just curious, not questioning to be a jerk or anything. I've never looked at the numbers that go into the rankings that much in the past, and now that I'm looking into them this year, it's pretty confusing how rankings/tiers are devised, especially when comparing pitchers vs hitters.

[mkultra]

A score of 13.406 for Johan and the bottom of the tier has Haren at 9.29. How does that work into "tier 1"? That seems like a huge drop considering the next 8 pitchers only have a difference of 1.533 between their z-scores. Also, without comparing all players (hitters and batters) across the board to each other, how would you start to determine what round you should be drafting these guys in?

Well, I'd personally put Santana/Lincecum/CC in a separate tier from the rest of those guys. Also, if you look at the Cafe tiers, it looks like they're using different projected stats than you, and also taking into account more than just W/K/ERA/WHIP.

As to "when to draft", that's where things other than your spreadsheet come into effect, IMO:

- When players fall below MDP
- When other managers start drafting guys in a given tier
- Your overall draft strategy
- Your gut

[BobbyRoberto]

RynoDyno,
First off, to answer you question from a few posts back, if you convert your z-scores into dollar values, then the pitcher/hitter value shakes out because you can choose how much to spend on hitting and how much on pitching (60/40, 65/35, etc.). Even if you don't do an auction draft, you can use the dollar values to help with a combined list. I'll try to explain this below.

That being said, I like what mkultra said about "when to draft." Once you get your rankings, you should have a column for Average Draft Position to see where your rankings differ with the average. Also, once you've established tiers, then you can adjust through the draft as far as when to start taking from which tier. It's usually best to draft at the bottom of a tier than the top of a tier. Tiers should have players close enough in value that you wouldn't mind having any of the guys in that tier, so whether you get the top guy or the bottom guy shouldn't matter.

Okay, dollar values. If you go back to your first post today (3/24), you had too many pitchers showing up in your top 10 because they have similar z-scores to the hitters. You can try this:

Figure out how much money would be spent on players in your league. Traditionally, the roto $$ amount per team is $260. You don't want to multiply $260 x 12, though, because you want what's called "marginal dollars," meaning the amount spent above $1 for each roster spot. So if you have 12 teams with 20 man rosters, then the marginal dollars per team would be $240 ($260-$20), and the total dollars spent would be $2,880 (12 teams x $240/team).

Then you can take that $2,880 and figure out how much should go towards hitters and how much should go toward pitchers. If you want to use 60%/40%, then multiply $2,880 by .6 for your hitters ($2,880 x .6 = $1,728). The remaining will go to pitchers, which leaves you with:

$1,728 for hitters.
$1,152 for pitchers.

Of course, some people prefer 65/35 or even 70/30, it depends on the person.

Next, add up all of the VORP scores for your hitters, then for your pitchers.

Take your total dollar amount for hitters (in this case, $1,728) and divide by VORP points. The result is the dollar amount per VORP point.

The final step is to multiply each hitters VORP by this dollar amount/per VORP, then add $1 (so you don't end up with players worth $0.).

You do the same with pitchers to get their dollar values.

Then you can combine the hitters and pitchers based on dollar values and should get a good list.

I hope that helps.

[RynoDyno]

thanks again for the help bobby, the $ values did the trick using a percentage towards hitters and pitchers. i had read that article you linked from tango before, but i forgot about it since i starting working on the rankings.

i have my list complete now, i took your suggestion about putting down each players mdp and then i have a column calculating the difference between their projections ranking and the mdp rank. two other things i also did that should help me:

1) color code each position, that really helped me see the tiers stand out

2) i took the avg of each column like SBpts for example, then subtracted that number from each players score in that category so while i'm drafting i can quickly see which player helps me most in the categories i need.

here's a screenshot sample of what i ended up with:



thanks again for your help!!!!

[BobbyRoberto]

RynoDyno,
That's a nice-looking spreadsheet you have there. I'm glad I could help.

So, if you were to have the 3rd pick in a 12-team draft, would you go with David Wright or Jose Reyes?

[RynoDyno]

RynoDyno,
That's a nice-looking spreadsheet you have there. I'm glad I could help.

So, if you were to have the 3rd pick in a 12-team draft, would you go with David Wright or Jose Reyes?

for my draft which uses ci/mi positions, not sure if i did this completely right, but i base these values on $value of each player minus the lowest valued player that will be drafted in the ci or mi slot, so i came up with:

$21.21 reyes
$21.12 wright

i personally like to have a balanced player like wright more at the #3 spot because of the solid #'s he puts up across the board, i feel i have more flexibility with later picks.

how do you rank them?

[Ender]

If for some reason I started a draft on auto draft and ended up Reyes as the 3rd pick knowing it was going to 20 more picks before I got to go again I'd be devastated. SB are easy to find, power is the rare commodity these days. No way I'm walking out of the early first round with 10-15 HR.

4th is the pick I don't want, I have Reyes listed 4th but I honestly can't see myself drafting him unless it is to try to trade him right away. Statistically he might be worth the 4th pick but the type of player he is generally aren't as valuable as the numbers suggest since you have to restructure your team to fit them into it. Minor leg injury or just a BABIP blip can easily derail his season since SB is his only real tool, give me a power/speed guy or a high AVG/power guy anytime of the week, those are the guys I want in the first half of the 1st round.

[BobbyRoberto]

My rankings have Wright slightly ahead of Reyes. Reyes' steals are appealing but, like Ender mentioned, a solid all-around guy like Wright would be tough to pass up.