June 19, 2007

Fantasy Focus

Projecting Final Standings

by Kenn Ruby

Printer-friendly

"Anybody know where I can go find some good projections?"

I lurk on a lot of fantasy baseball message boards, and this request comes up over and over again all winter long. In December and January we just want any projections--good or bad--but the closer we get to Opening Day the more we need to find reliable projections. For fantasy players, it's our Holy Grail.

Even the best prognosticators in the business fall on their faces every now and then. Case in point, RotoWire predicted Alex Gordon to have a .293-24-102 triple crown stat line; his current pace would put him at .215-12-39. For the most part, we're in the ballpark, however. Isn't that really all you need?

In my NL-only draft, I bring spreadsheets that have the projections of the one Nostradamus I trust. If I know that a player is projected to hit 24 home runs, I generally can expect he'll hit 15-30 home runs. On draft day, it's nice to just see that there's a guy capable of hitting 24 home runs still available. I don't need to know that he'll hit exactly 24, just that he's at that level. You probably do something similar.

The thing is, it's the middle of June now. The standings in your league reflect real numbers--not projections--and if your league is anything like mine, there's been plenty of roster turnover. Are those projections from last spring meaningless? Hardly. The key to your league may still be inside those numbers.

Most teams have played around 70 games this season, meaning the season is about 43% over. You can easily project the rest of the season by taking the preseason projections and pro-rating them to 57% of the season. Why not? If you trusted the projections three months ago, why wouldn't you do it now? It's easy enough to do on any spreadsheet, and you could make adjustments depending on the number of games played by any team at any point in the season.

How to Calculate Projected Standings

To calculate projected standings, you would simply add the year-to-date totals to the projected totals for all players on active rosters. With the aforementioned roster turnover, your year-to-date league standings may not reflect the current rosters. For example, due to the numerous injuries my team has suffered, I recently had Paul McAnulty and Ryan Ludwick in my active lineup (combined projected at-bats before the season: 30). This morning's standings may have told me that I was a contender, but when I project everything, it's clear that my team isn't built to go the distance.

Some might balk at such a crude method. You might point at breakout stars and say, "what good are those projections?" After all, Prince Fielder already has 24 home runs, and he was only projected to hit 30. The danger would be misinterpreting the projections to assume he's only going to hit about six home runs the rest of the way. Really, he's projected to hit about 57% of 30 the rest of the way (since we trust that the projections in March represent his true skill level). That means he should hit about 17 more bombs. That 11-homer difference could be the difference between first and second place in your league.

There are drawbacks to using the projections this way:

1. Serious injuries can skew the results. Chris Carpenter is projected to get about 10 more wins and 100 more strikeouts. That isn't happening.
2. While projections of skill don't necessarily change, playing time and/or a player's role can change dramatically, especially with closers. Salomon Torres/Matt Capps, Brad Lidge/Dan Wheeler, and Brett Myers/Tom Gordon probably don't have the most accurate save projections any more.
3. True breakouts--especially from unpredictable rookies--often get missed. Mark Reynolds is a great example. He's come down to earth some, but he's still on a lot of rosters. Consulting my draft day spreadsheet, I just noticed he wasn't even on there.

Still, most seriously injured players aren't on active rosters anyway (in other words, Carpenter's 10 wins would be irrelevant in this exercise), and there are so few instances of the other two problems that you could easily make a few manual adjustments after projecting your standings.

Once you have your projected final standings, you might notice a few things. For one, the team projected to win might not be the team in first place right now:

Year-to-date standings

          R     HR    RBI     SB    AVG     W    SV    K    ERA    WHIP   TOTAL
-------------------------------------------------------------------------------
Team A    11    11    10      10     11     5     8   11      9      8      94
Team B    10     5     9      11     10     2     6    6     11     11      81
YOU        9     3     7       5      6    11     9    9      5      9      73
Team D     8    10     2       4      4     7     5    4     10     10      64
Team E     5     7     4       2      8    10    10   10      2      2      60
Team F     4     2     5       7      9     9     1    7      7      5      56
-------------------------------------------------------------------------------
Team G     6     4    11       9      3     6    11    3      1      3      55
Team H     2     8     6       6      5     3     7    2      8      7      54
Team I     7     6     1       3      7     8     2    5      6      6      51
Team J     3     9     3       8      2     4     3    8      4      4      48
Team K     1     1     8       1      1     1     4    1      3      1      22

Projected standings

          R     HR    RBI     SB    AVG     W    SV    K    ERA    WHIP   TOTAL
--------------------------------------------------------------------------------
Team B   10      6     10     11     11     5     9    6     10      9      87 (+6)
YOU      11      5      8      9      7     9     8    8      8     10      83 (+6)
Team A    9      8      9     10      9     7     6   10      6      8      82 (-12)
Team E    7      7      5      4     10    11    11   11      2      2      70 (+10)
Team F    8      2      7      7      8    10     1    9      5      4      61 (+5)
Team D    4      9      2      2      4     4     5    4     11     11      56 (-8)
--------------------------------------------------------------------------------
Team G    6      4     11      5      5     6    10    3      1      3      54 (-1)
Team H    2     11      6      6      3     2     7    2      7      7      53 (-1)
Team J    3     10      4      8      2     3     3    7      4      5      49 (+1)
Team I    5      3      1      3      6     8     2    5      9      6      48 (-3)
Team K    1      1      3      1      1     1     4    1      3      1      18 (-4)

Team A may be feeling overconfident with that 13-point lead. He's certainly not thinking of you, 21 points behind. The projections right now say that Team B is the real threat, and that they could overtake Team A with their current roster. Imagine if Team A gets complacent and Team B gets desperate? Armed with these numbers, you can swoop in and make a trade to gain yourself a few points (look at home runs in particular) at the expense of your closest rivals. Always keep your eye on Team B--he's likely to make adjustments as well. Meanwhile, Team A might think things are going fine and do nothing.

Secondly, your projected standings might reveal areas where you can gain without making too much of an effort. Perhaps your year-to-date standings show you 20 stolen bases behind your nearest competitor, but your projections show you only a couple of steals behind a group of three. Trading for one stolen base guy could gain you three points easily, and even if you stand pat you could still catch that group--an important lesson if you're thinking of breaking the bank.

Third, those projections may expose weaknesses on your team that you haven't recognized yet. Perhaps you rode the hot starts of three or four hitters. Once the projections start to become fact, you realize that maybe your offense could actually use another bat, because there's no way your early-season heroes will all keep up their current paces. Maybe that 3.25 ERA is not sustainable with the rotation you have. Maybe Brad Penny just isn't that good.

Fortunately, with your projections in hand, you might trade the Brad Pennys of the world for someone who can really help you; Penny's 2.18 ERA could net you those three stolen base points. You always run the risk of getting burned--maybe Penny is that good--but if you truly trust the projections and can see where the season is going, you have to take the chance.

You might think doing these projections is time-consuming. It will take you a few hours to do it all, particularly the first time you set it up. Once you get used to it, however, you'll find that the time spent will go by quickly, and you'll be able to adjust the work you'd done previously each time you run your projections; I'd recommend every three weeks or so. It could be that the "work" you put in right now could be the single most valuable part of bringing you that elusive title.