|Target Drafting: Not All it is Cracked Up to Be|
|Chance Favors the Prepared Mind|
|Written by Todd Zola|
|Tuesday, 15 February 2011 02:14|
One of the more popular strategies employed is what I call target drafting. The idea is to look at the history of your league and determine what it takes in each category to win. On the average, you need 75-80 percent of the total points in rotisserie style scoring. In a 5x5, 12-team league, that total comes to 96 points, or third in six categories, fourth in the other four. Most people just shoot for third place across the board. In 5x5, 15-team leagues, you need 120 points or fourth across the board. So what you do is determine what is needed historically in your league to finish at that spot and draft towards that target as your category goal.
I admit, for several years I adhered to this means of assembling my teams. I tracked my category totals via projections. Some even credit me with being one of the first in the industry to publicly write about the concept and I was definitely one of the first to provide targets for common leagues. I went so far as to detail a means of adjusting your targets based upon the projections you are using. And now, I am going to be amongst the first if not the only person to publicly challenge this notion and question its utility.
But before I begin, the take home message is not going to be suggesting you cease target drafting. If it affords you comfort and you feel it aids in your construction of a championship team, by all means, nail those targets! It is just that I am now of the mind that if you indeed are victorious it is not because you hit your targets. But then, it is not in spite of it either. I am not about to contend that target drafting is necessarily detrimental, though I do feel there are a couple of possible pitfalls associated with the process.
Here’s the deal. Ask yourself a couple of simple questions. Since you have begun target drafting, how many times have you failed to meet your goals? And how many times have you won leagues in which you did hit your targets? I am guessing that almost every time, you left the room with a squad you feel “won” the draft or auction. That is, if you added up the totals per category and scored it roto-style, you not only met your goals but finished atop the mythical standings. However, come October, your record may not be so spotless.
Now ask yourself another question. If you polled your league mates and asked them how many of them made their targets, would you be the only one? I doubt it. In fact, I bet pretty much everyone will brag they nailed their goals. But how many of you ended up with the proverbial Yoohoo shower? Hmmm.
There are several issues with target drafting. Some are more anecdotal but others can be demonstrated quite readily using some interesting historical league information.
Let us start with the fact that the process is seriously skewed due to the bias in the projections you are employing to put together your team. You are basing your selections off of projections you favor. You are picking players you expect to perform well and leaving the rest for your opposition. And since not everyone evaluates and ranks players the same, your category totals are going to be artificially inflated in your favor. This example is complete hyperbole, but it illustrates the point. If you predict Willie Bloomquist to hit 25 HR and steal 35 bases, chances are you are going to meet your targets. Again, please do not get lost in the absurdity of this example. The point is, if you pick 14 hitters that you feel will hit 2 or 3 more HR or steal only 2 or 3 more bases than everyone else does, you are now +30 in each category and will likely meet your objective. The problem is, you may have your team pegged at 280 homers and 180 steals, typical benchmarks to finish third in 15-team leagues, but everyone else in your league has your totals at 250/150. Similarly, someone in your league may proclaim they drafted 285 homers and 160 steals and you may have them for 240/130, or whatever. What I am saying is you sure as shootin’ better meet your targets and “win” your draft, or there is something wrong with your drafting process.
Another issue I have with target drafting relates specifically to leagues with reserve lists, especially with respect to pitching. You are not playing in a draft and hold league. You will be making weekly and sometimes more frequent changes to your roster. There is no way to account for this roster management in category targets. Chances are, you will be benching your starting pitchers with difficult matchups and subbing in those with more favorable matchups. In theory, this should help your ratios. Also, you will be deploying two-start pitchers in an effort to boost up your wins and strikeouts totals. Again, this is impossible to capture since you are tracking the entire season expectations for pitchers you will be benching and not counting the stats of those active in their stead. To me, this makes the notion of meeting pitching targets almost moot.
But perhaps the strongest reason why I have grown to dismiss the practice in my leagues is at some point, at least a third of your roster will be spending time on the disabled list, to be replaced by what is usually a lesser player. In addition, most leagues allow trading. The end result is the players you have active on opening day are only going to provide you a portion of your year ending totals. How can you judge if you have ample stats to win when you have no clue who is going to play and who is going to be replaced?
To that end, I looked at a league I played in from last season. I took the opening day lineups and calculated the year ending totals assuming it was a draft and hold league. I then compared that total to the real team totals, after 26 weeks of roster management. On the average, about 66 percent of the counting stats in the hitting categories were originally drafted. On an individual team basis, the spread was from the low 40s to the high 80s, with no significant correlation between a team’s percentage and success. That is, a large part of each team’s ability to win revolved around their roster management. Looking at the pitching, 80 percent of the wins were drafted, 81 percent of the strikeouts and a surprising 86 percent of saves were on a roster as of opening day. As an aside, do you still think waiting to pick up saves in-season is smart? And as another aside, this data serves to further back my contention that one should focus more on pitching than in previous seasons, as there is not as much help in the free agent pool anymore.
Let us get back to the topic at hand, target drafting. If on the average, you can expect to only accrue 66 percent of your target with error bars of 20 percent, does meeting your goals really insure victory?
Putting the numbers and other arguments aside for a minute, a fundamental issue I have with target drafting is I fail to see the difference between drafting a player you expect to give you the most homers so you can attain an arbitrary, sorry, I said let’s put the arguments aside for now, so you can attain a preset goal and just drafting the player you think will hit the most homers? My point is, I don’t see how most choices would be any different whether or not aiming for targets is involved. Your goal is to draft the players you feel will give you the most stats. Who cares how many.
Now, I realize that I am setting myself up for the argument that tracking stats, especially homers and steals can help keep a balance between speed and power. And, if you say you are going to continue to track for this sole purpose, who am I to argue. But, there are a couple of dangerous pitfalls you need to avoid. We are treating the projections as static numbers. They do not reflect the range of probabilities associated with performance and/or health risks. Maybe the best way to make my point is to offer a couple of examples. Let’s say you are 20 homers shy of your target. Available are Jack Cust and Marlon Byrd. You figure Cust for 25-30 bombs and Byrd for mid-teens. Ignoring all the other categories, which is the better choice, a sure thing 15, leaving you 5 dingers short or a risky 25-30, putting you over by 5-10. When I consider the fact that I am going to need to find about 34 percent of the homers I need to win during the season, I personally prefer the more stable production of Byrd over the volatile Cust. If you happen to be a Cust fan, sub in the name Russell Branyan last year and Richie Sexson in years before that. The point is, you do not want to be drafting a static number just so you can leave the room over your targets. Another example is that you may be short of your stolen base objective, but have already made your other numbers. Do you take Ryan Theriot just to make your stolen base target or do you take a more rounded and more valuable player and look to find steals in season? I know what I would do.
As you can see, I am not going to campaign that we outlaw target drafting. But I did want to bring to your attention that the concept is only a tool, and perhaps not as useful as you previously thought. If you are aware of the dangers but are still comfortable with the guidance that drafting towards targets may provide, knock yourself out. But, if you tell me you drafted 275 HR, 1110 runs, 1076 RBI, 154 SB, .281 BA, 87 wins, 3.45 ERA, 1.232 WHIP, 1018 K and 76 saves and the standings have you winning the league, be ready for an eye roll followed by “really?” And if you end up winning, it is not because you drafted those totals but rather because you did an excellent job of supplementing what you did draft during the season.
Before I leave you, I do want to mention that Platinum subscribers will be getting a more complete treatment of this topic which will include access to a spreadsheet detailing the results of the league discussed above along with similar breakdowns of a few others.
|Last Updated on Tuesday, 15 February 2011 11:06|