As someone who’s basically fanatical about his fantasy sports, I’m often asked by people if they could have my draft lists. They figure "he knows what he’s doing and he spends a ton of time doing it, so of course whatever he has is going to be better than what I can put together." While some of that is, of course, true (my back enjoys the self pat), the reality is this is an inexact science.

The explosion of the internet has left fantasy owners with tons of information. You’ve gone from site to site and downloaded draft lists. But more importantly, you’ve downloaded a spreadsheet full of projected stats. So now what? Do you just stare at it until something makes sense? While many do just that, this article will give you a brief introduction on how to turn those projections into player rankings. I’d be remiss if I didn’t give a quick shout-out to by my boy CA who introduced me to this way of thinking.

First, let us consider the home run. While opportunity, and by that I mean a higher at-bat total, is often a key factor into creating someone’s home run projection, the reality is you don’t care if a player hits 25 homers in 500 at-bats or if he does it in 350 at-bats as part of a platoon. A homer is a homer. So for valuation purposes, let’s just assume each home run is worth one point.

Second, what about runs and RBIs? These categories are a bit trickier. While it’s true that you don’t care about at-bats for these counting statistics. The fact is, without enough plate appearances, you’re just not going to get enough chances to knock in or score runs. So, while some will reduce the importance of these categories in their projections and in their valuations, I take a slightly different approach.

I like to assume that a run and an RBI are similar in value to the home run. In fact, a home run will also result in a run and an RBI so they are somewhat connected. The difference is, while a home run is purely the byproduct of the player, runs and RBIs have a dependency factor. Outside of the home run, you need help from your teammates in order to generate results. So, how do we square them up closely? First, based on historical data (and you can check your leagues historical standings if you don’t believe me) runs and RBIs occur at very similar paces

** SMALL SAMPLE SIZE ALERT:** In looking at a 20-team league I run; In last year’s standings there were 17,599 runs scored and 16,814 RBIs. That’s a difference of 785. Since the league is 20-teams, that’s 39.25 runs per team. Consider 14 starting offensive spots and that’s 2.8 runs per person more than RBIs. Convinced they are somewhat equal? What I then do is add runs and RBI together. In order to convert it back to something that makes it on par with homers, I choose to divide that number by some factor, say 3.8. So, a player projected to score 100 runs and drive in 100 gets a raw score of 52.63 (200 divided by 3.8). That’s roughly 26.31 points per category. So 100 runs and 100 RBIs are the equivalent of hitting 26.31 homers. Therefore, a player with a projection of 26-100-100 will now have a raw score of 78.63. We’re on our way.

In the next installment I will discuss how to factor stolen bases and batting average into the mix to complete your raw scoring.

Jeremy says

It seems that you are randomly choosing a number to divide your combined RBI/runs total. “I choose to divide that number by some factor, say 3.8.”

Why 3.8?

And why, after factoring in the projected HRs is the total 72.63? Where do you get that extra 20 points?

If 100 runs + 100 RBI = 52.63 points (which also equals 26.31 HRs), wouldn’t the total of RBI+RUNS+HRs be:

52.63 (100 RBI + 100 Runs) + 52.63 (26 HRs) = 105.26?

Jonathan Stulberg says

3.8 is not random. You want to scale runs and RBI so you’re not valuing them more than home runs. If you used 4.0 for example 100 runs would be the equivalent of 25 homers. I choose to use less than 4 because 100 runs is worth slightly more than 25 homers. As for the math, if you remember a homer is only worth 1 point. So in the example the value of the RBI + runs is 52.63 (which when you divide by 2 is roughly 26.31 per category). Then when you factor in the 26 homers at 1 point each you get 78.63 (52.63 + 26). Thanks for the comment since you pointed out a typo in the final score.

Jeremy says

Ahhh, that makes more sense. A typo. Why does the simplest solution never occur to me first? I think choosing 26 homers as an example may have also confused me a bit just because the number is so similar to 26.31. But thank you for the clarification! Looking forward to the next installments.

Jeff says

If I was going to critique the methodology, I’d recommend you calculate a factor that better reflects the relative value of runs and RBI compared to HR. 12 players scored 100 runs last year, and 19 had 100 RBI? Compare that to 39 players with 26 HR. By comparison, the 39th place RBI total was 86. A more useful factor for RBI would therefore be 3.3. As your formula stands now, you overvalue HR compared to RBI by almost 15%. The math for runs scored is almost identical.

Jonathan Stulberg says

Jeff. Thanks for the great comment. There are a few things to take into account. First, I want HR to be a bit more statistically significant. Home runs are an individualized skill. Usually when a player owns a certain propensity for HR, he maintains that. RBI and Runs are a dependent skill. You must rely on your teammates. It’s also a category where you’re more often than not going to be more significantly off on your projections so you want to make sure you account for that. Secondly, HR (and SB) are a category where you’re going to see a huge disparity amongst players. A guy like Michael Bourn isn’t going to come close to Miguel Cabrera in HR but they can be statistically similar in runs scored. Therefore, you want to give a slight extra boost to HR (and SB) over runs and RBI. Finally, and I think this is the most important point, is I wrote this article and gave you the proprietary formula because I want people to use their own judgment. If you’re more comfortable weighting RBI and Runs a little higher, than change the formula to 3.3. In Part II of this article, you’ll see where I give you the formula for calculating batting average influence but I tell you to use your judgment based on your league parameters in determining what baseline average to use. My article isn’t intended to give you fish. I’m trying to teach you to fish. Tweak my formula to meet your personal bias and preferences and see how that goes.

Jeff says

Jon, Overall I like your approach. I’ve been fishing for years and have developed my own, which is quite different from yours, but to each his own, right? My only other critique would be that your method lumps all stats into a single number, which could lead you to over-draft certain stats and underdraft others. For example, if you blindly follow a single-value rating system, you could end up with a team of Jose Reyes, Michael Bourn, etc. and win SB by a margin of 150 and finish dead last in HR and RBI. Not a successful season. Basically, your single-value system treats all incremental HR (and RBI and SB, etc.) equally, when in reality the only value is within a range. If 300 HR gets you 1st place in your league, there’s no value to your team’s 301st, 302nd, 303rd HRs. Similarly, if 220 HR is the last place number, it doesn’t matter if you have 220 or 100. You’re still last.

Jonathan Stulberg says

Thanks for the reply. My calculation is purely for the creation of your draft lists. It’s a measure of what a player’s “value” will be should he actually meet these projections. It’s not meant to be an indraft calculator. That’s a completely different animal. If you read Part II of the Value series, you’ll see where I state that during a draft, you still want to be mindful of certain categories and that you should draft accordingly and simply choose between like-ranked players. My example was that if you drafted Michael Bourn early, there’s no need to take Brett Gardner a few rounds later simply because he’s the highest rated player on the board. You take the power hitter who’s close. For example, if Gardner’s score was a 92.35 and some power hitting outfielder was 90.02, you’re better off with the 90.02 hitter because you already have Bourn. I’ve tried to use the “draft programs” that people have offered and quite frankly I found them to always steer me in a direction different from the way I want to go because they assume everyone is going to match their projection. I prefer to look at my projection more objectively and will sometimes zig when a draft calculator would tell me to zag. If you want, you can email me and we can talk more about your application of the strategy.

Beav says

In order to make a decision between the speed oriented guy and the power heavy guy who are evenly rated, keep a spreadsheet to drop your drafted player stats into as you pick them. Use two spreadsheets, one split into hitting categories and one for pitching categories. Each column of the spreadsheet represents a category. At the bottom of each column, place your goal for each hitting and pitching category that historically would put you at the 80th percentile for that category for your league. (ie. 16 points in a 20 team league)

With each pick, enter that players stats into the spreadsheet with the spreadsheet recalculating cummulative category totals with each selection input so you can monitor your squad’s cumulative stats against your goals. As you make picks, be careful as Jeff said earlier not to overload a category so that your 200th steal cannot possibly benefit you when it comes at the expense of your putridly low homerun total of 150.

Nail your projections and you score a 160 out of 200 in a twenty team 5×5 league, usually more than enough to take home a title.

Clint says

I guess this article is working on getting to answering this fantasy baseball question for me but since I’ve been playing since the early 2000’s and not won a championship yet I’ll ask: what categories are hardest to back fill via trades mid to late season, i.e. catch up on? I’d presume this leads into draft strategy but like I said, I’m still trying to figure out a winning formula one way or the other and haven’t toyed with the idea of concentrating on getting way ahead early on and then trading for a late run at the title so I’m curious if it’s worth looking at. Typically, I’ve drafted for well-rounded teams but the closest that’s gotten me is 2nd place twice (one time losing 1st place in the last 48hrs of the season) so I’m open to new strategies.

Jonathan Stulberg says

Thanks for the question Clint. Generally, Wins are the hardest to back fill for a number of reasons. First, late in the season, there’s only so many starts left for the more reliable starters in the league. Secondly, as we all know, pitching great doesn’t necessarily turn into wins. Finally, most league have a limit on innings or starts so late in the season it’s tough to pile on. The other category that’s hard is AVG. You’ve already accumulated so many at-bats, getting a .300 hitter late in the season doesn’t make that much ground up. That all being said, it really depends on your league standings. I’ve been in leagues where .002 in batting average meant like 5 spots in the standings so every little bit helps. Personally, I take your approach. I try to draft a well balanced team. Of course, it never works out that way so the big key is DO NOT FALL IN LOVE WITH YOUR PLAYERS. If you have too much speed, don’t be afraid to trade it early, even if it’s a player you really like. Too many guys fall in love with the players they draft and ask for too much. Don’t be afraid to deal, even in May. That should help you get over the hump and win a title.

Gene Krywonis says

I don’t see how you came up with the values for PITCHERS….anywhere…Kimbrel at 160+ ? Explain that one especially…..The HITTERS categories are explained very well in the 2 articles…To all the illiterates out there who don’t understand “value” keep re-reading a few times ..it’ll eventually sink in !

Jonathan Stulberg says

Gene. Thanks for the question. For the pitchers, there’s a number of factors that go into it. For starers, I take my projected win total and multiply by .75. Since Wins are a crapshoot and hard to predict, I downplay their importance. I take my K projection and divide by 4. Yes, a 200 K pitchers gets 50 points for his K total but a 20 game winner only gets a 15 for his wins. Again, wins are not predictable and quite frankly, a high K pitcher will more often also get his share of wins. ERA and WHIP are where it gets tricky. Again, I think the baseline you need to figure what works best for your league. For my preference, I think a pitcher with a 3.95 ERA is neutral. Obviously if you play in a 12 team league, you’ll lower that total. My leagues are 20 teams. Then I take 3.95 and subtract my projected ERA. Take that total and divide by .03. So a pitcher with a 3.05 ERA will get 30 additional points ((3.95-3.05)/.03)=30. Same idea for WHIP except my baseline is 1.29 and I award a point for every .01 difference. Since RP very often have much lower ERA and WHIP, their scores tend to be more inflated as I haven’t yet figured out a comfortable way to adjust that downward. In addition, because the scoring systems aren’t scaled, we rank them separately so as not to suggest Kimbrel as the #1 overall pitcher. This year there was some influence based on innings pitched (because as we all know, the more innings, the better the benefit or harder the detriment will be to ratios). It’s still a work in progress (and I think it will be forever because projecting and ranking is an inexact science anyway) and hopefully it at least helped you rank players for your drafts. With each year we will hopefully hone our craft even more.

Jonathan Stulberg says

One more comment. For RP I use saves instead of wins as that multiplier which once again explains how RP have a higher scaled score as saves are more plentiful than wins.