What to make of pre-ranks part 2

In last week’s article, I looked at how Yahoo’s preseason top 50 have performed this season. Not surprisingly, the overall reliability of the rankings were low and it’s logical to presume the the reliability of the rankings from 51-100 would not be better.

For me, many of the toughest draft or keeper decisions arise when evaluating players in the 51-100 range. By that part of the draft, the sure things have already been taken, so managers often use pre-rankings to differentiate between players. So, let’s take a look at how Yahoo did predicting the performance of this segment of players.

Player Type	Pre-ranked 51-100 (PR51-100)	Actual 51-100 (A51-100)	PR 51-100, not A51-100	Not PR 51-100, A51-100	PR51-100 vs. not PR 51-100 as % of A51-100	Success Rate
Batters	32	30	18	16	47/53	44
Starters	12	11	8	7	36/64	33
Relievers	6	9	3	6	33/67	50

So, what conclusions can we draw from this?

• Acknowledging sample size caveats, elite relievers were a safer bet than the starters or hitters in this group. On the other hand, elite relievers have lower ceilings: Jonathan Broxton, the highest ranked reliever this season, is currently the 38th most valuable player overall. (Broxton wasn’t in Yahoo’s top 100 pre-ranking.)
• Eliminating relievers for a second (due to their low upside), 10 out of the 16 players who lived up to their top 100 ranking exceeded their pre-rank position by more than 25 slots—or about two rounds. The takeaway here should be that you shouldn’t be timid about drafting a player “early,” or keeping a player when you have others who may be ranked above that player. There are very few sure things in fantasy baseball.
• Only one-third of pitchers pre-ranked in the 51-100 range delivered top 100 performances. This once again affirms the idea that it is a poor idea to over-invest in starting pitching. A viable strategy is to get one elite pitcher, one pitcher from this group, preferably one who either slips or who you feel is undervalued, and then load up with high-upside depth later in the draft.
• Stolen bases are reliable value. Six of the 13 offensive players who held their value have either stolen, or are on pace to steal, 20 or more bases. Of players who are on pace to steal 20 or more bases, only Nate McLouth and Vernon Wells won’t justify top 100 pre-rank status. As the old cliché goes, speed doesn’t slump. That is a good thing for fantasy value because each individual steal is more valuable than a single run or run batted in.

Finally, I want to expand a bit on a disclaimer I made in the comments section of the first part of the pre-rank analysis. I don’t really know what a laudable success rate would be for pre-ranking. A 42 percent success rate when picking the top 100 players may actually be very good. Further, this analysis was somewhat crude, and there are many alternative ways to evaluate the pre-rankingss. For example, I could have analyzed ranking by position, or used “within 25 slots of the pre-rank” as the criteria for “success.” The greater point of interest is that it seems there hasn’t been much of a formal movement to promote accountability for pre-rankings by fantasy heavyweights like Yahoo.

The goal of this (mini-) series was to give fantasy baseball players perspective on the amount of credence to pay to pre-ranks, not to bash Yahoo’s performance. Because fantasy baseball is very unpredictable, it would be unfair to judge Yahoo’s pre-rankings based solely on success rate; as long as the rankings are independent and based on sound reasoning, they can add value for fantasy players. Sites like The Hardball Times may not always be correct either, but we do strive to meet the same standard of sound reasoning and independent analysis. And with that, it seems like a perfectly opportune time to plug The Hardball Times Baseball Annual 2010 because that’s the type of content you can expect from the good folks here.