Drafting and the Slingshot

I was just thinking a bit about the recent UPL draft, which I've recapped in some previous posts (also, Greg's put up his 2 cents worth). And I've come to a conclusion.

This was the highest caliber draft in the history of UPL Baseball. We've brought back most of the familiar faces, with Black Sox now a UPL fixture, CJ (JimmyDix) coming off of a 4th place finish, and a new, (presumably knowledgable) franchise in Hats for Bats. Naturally, strategy was going to be important.

A while back, I posted about draft strategies. An excerpt (with emphasis added):

The first place that the metagame matters is in the UPL is in the draft room. The first rule should be obvious. Figure out who's auto-drafting based on Yahoo! rankings. This is the single most important aspect of drafting in online drafts where not everyone is there (like in the UPL). So think of it this way. If you were the only live drafter in a draft, with everyone else auto-drafting, then you know exactly what players will be available for you. This is a huge advantage, as you turn a very dynamic problem, with 12 individuals, all acting independently, into a very static problem, with 1 individual operating in a known environment. We've never had a draft with only 1 live person, but we have had scenarios with multiple auto-drafters. A dream scenario for me would be something like drafting 8th, and having teams 9, 10, 11, and 12 all on auto-draft. In this scenario, I know with some precision exactly which 8 players are coming off the board before I draft again (obviously, this only works in one direction on the S curve, but you get the idea).

One thing that I immediately noticed was that I had a horrible draft position, based on who else was drafting live. Obviously, you want to be surrounded by auto-drafters (Westy, Benver, and Cheeseheads, if I recall correctly). Failing that, you want to be surrounded by predictable drafters. Of the UPL folks, you can put OD (likes the Red Sox), CJ (likes the White Sox), and Rupert (likes young, high potential prospects) on that list. And to a lesser extent, Westy (likes Minnesota).

Take a look at the top 4 picks. C-Lauff, Black Sox, Hats, O.N. Thugs. Defending champ. Strong UPL Newcomer. UPL rookie, but knows baseball (we think). GOAT. From my standpoint, I've got 3 of the top competitors in this league all drafting in a row, so that there are 6 picks in a row, where I can throw out the pre-ranks. Absolute nightmare of a draft position. The only worse position? Mush Heads, who also had me to worry about. In any case, this situation is basically the opposite of what I had written about as being ideal (having a bunch of autodrafters bunched up before/after you).

Conversely, look at the 9-12 slots in the draft: OD, Westy, Rupert, Benver. If you're drafting 8, then you know that after you pick, the next few picks will be: Best Red Sox Player Available, AutoDraft, Reach For Young Prospect That Won't Matter for 3 Years (If Ever), AutoDraft. And if you take a quick peek at the 6 and 7 slots, you have Cheeseheads and OD. So, you'll know that the players selected will be: AutoDraft, Best White Sox (or AL Central) Player Available. Greg had the fortune of being in the 8 slot. As I think about it, I wonder if I wouldn't have been willing to trade slots with Greg. Obviously, drafting at 4 give you a better 1st round pick in a somewhat top-heavy draft. But that 8 slot is really, really attractive.

In any case, based on what I wrote before, you'd figure that I'd be screwed. But I found one interesting thing associated with drafting at the end of a line of live drafters. In the early-middle rounds, I was able to be at the start of the runs that were made at the scarce positions. I'm going to call it the slingshot, like in NASCAR where you use the aerodynamic edge from the line of drafters to give you a little momentum coming out of the turn. In terms of the draft, you use the trends that a line of drafters builds, and then when you change directions, the effects are more dramatic.

Round 4 - Bottom of Top Tier SP. Available: Peavy, Halladay, Liriano, Billingsley, Haren.

Round 5 - Top Tier C. Available: Martin, McCann, Mauer, Soto, Victor Martinez.

In both rounds, I was able to get the guy I wanted. But more importantly, as I'm going into the turn, I'm able to sort of set the pace in the 4th. But coming out of the turn, I can really change directions, which I did by going C. The result is that the folks in the 1, 2, and 3 slots didn't really have the chance to respond, and one of them made a reach in the 6th (Ianetta) as a result. Now, something a little different came about in the 8th round, but it was set up in the 7th round.

Round 7 - Top Tier RP. Available: K-Rod, Soria, Jenks, Rivera, Nathan, Lidge.

Round 8 - Second Tier RP. Available: (Soria), Valverde, Broxton, Marmol, Capps

Round 9 - Second Tier SP. Available: Harden, Volquez, Burnett, Price. Not available: Joba, Kazmir.

I was able to get the two guys that I really wanted in the top tier of RP. But how it happened was interesting. I want to go closer in the 7th, so I go K-Rod, and gamble that I was the only who had Soria in the top tier of closers. The run on the top tier closers runs close to expected in the 7th, and on the way back, in the 8th, I get Soria. This puts pressure on C-Lauff, Black Sox, and Hats to go closer. My plan was to go SP in the 9th, and I was hoping that Kazmir or Joba would be available, because everyone was reaching to closers. It almost worked, but Black Sox held firm. But I was still able to get Harden (who I had behind Kazmir, but ahead of Joba).

Overall, when I was heading into the live drafters (4th and 8th rounds), we saw a response, where they pretty much snapped up the rest of the group. And when I was going away from the live drafters (5th and 7th rounds), they often didn't have a chance to respond, until later, at a somewhat inflated cost (I wouldn't really want to go into the 2nd tier of closers until the 10th round, but the teams were left scrambling, more or less).

Of course, the draft order doesn't matter nearly as much once you hit the 15th round, when everyone's guessing. The odds of someone guessing the same as you is much lower than in the early rounds.

I think that what I figured out during this year's draft has changed how I view drafts. In the past, my strategy was to use the full listing of player ranks, within each position, make a few edits, and then to draft based on value, with an emphasis on position. I think that new strategy is to create tiers of players based on similar value, rank players within a tier, go after the guy that you want, and then see if there's value still available in that tier. This does a couple things. First, it gives you a more concrete view of how valuable you think players are, which in turn makes it easier to see what a drop to the next tier will look like. This gives you a better feel of when to try to start a run on a given position, and when to participate in that run.

In retrospect, it looks like it may have been a bit tougher for the folks in the 1, 2, or 3 slots drafting, since they weren't able to participate in the runs (unless they were willing to reach for a pick, like one of the top tier C's in the 4th round).

In any case, I think that our view of drafting is getting more sophisticated, especially as we try to get a leg up on an increasingly skilled environment. Of cours, this may be easier if I'd actually prepare before hand, and not try to do this on the fly during the draft.

-Chairman (aka O.N. Thugs)