The SSDS Drafting Method: Part 1 of 3
"The System"
published December 13, 2006

Last night, I participated in an invitational mock draft hosted by MockDraftCentral.com.  I told readers in my blog earlier in the week that I planned to use a still-experimental drafting strategy that I had used to win a real 20-team mixed league once before, in 2005.

This will be a three part series in which I will (a) show you the system, (b) take you through my thoughts and picks as the actual draft proceeded and then (c) show you exactly how the team shapes up, beyond what you might notice by glancing at the projections.

With the mock draft out of the way, I can tell you exactly what the strategy is and how it works. Whether it will be a successful one in the long run is too early to tell as I believe this is only the second real draft I've tried it in.  I have run many draft simulations on it and I feel that it produces much more of a consistently contending team, perhaps at the expense of the best possible average result.  In other words, despite winning in the only real league I've used it in, my theory is that it produces reliable second place or third place type teams with a tighter margin of error and less of a chance of going for the glory.  This is a good step if you're a player who has had trouble ever finishing in the top three spots and it may actually be better than that but we're just not far along enough to know yet.  Take this as your final warning that the system remains experimental.

I will not give into the temptation to summarize so-called projected standings from last night's draft based on current projections. Such an exercise is usually a self-serving one as each of us who draft use a different set of projections and naturally, our own team's place in the projected standings will benefit from being boosted by using the very numbers on which we base our drafting decisions.  I'll leave it, therefore, to one of the other participants to claim victory.

For lack of a better descriptive name, I've decided to call this the "skip to the safe" approach or, for short, the SSDS or "Skip to the Safe Draft Strategy."

At least a couple of observers wrote me late last night to tell me that my draft went very much against what our online ranking sheets would produce. In fact, as I'll show you here, last night's draft was not only quite mechanical under this system, the drafting strategy was actually formed almost entirely from the same ranking sheets that subscribers to our site have. It's just the way I use them under this strategy that makes this approach a bit different.

The philosophy behind SSDS includes several main goals and/or benefits:

(1) It is designed to avoid even slightly risky players or projections, including most rookies.

(2) It is designed to maximize perceived value for later trading (which obviously doesn't mean much in a mock draft but can be everything in a real one).

(3) It removes most of the emotional attachment we have to specific players or projections.

(4) It's relatively easy to use in a real draft and come away with a competitive team that should be in contention, even if it doesn't necessarily maximize projected value.

To execute the strategy, we first have to create a ranking sheet based on the number of fantasy teams, categories being used and so on. I did this using the member pages of our own website, highlighted the entire ranking chart exactly as the fantasy domination ranking sheets produced and copied and pasted the entire table right into Excel, as I often do. This allows me to then add notes or keep track of players during a draft by filtering rows as needed and so on. I only need to copy and paste the top section of the fantasy domination sheets, the part that ranks players in descending order of projected value. Using Excel's search feature, for an online league, I really don't need to have the sections that list catchers on their own and so on.  It's just as easy to use Excel's filter function to find the next highest remaining catcher and so on (e.g. Excel's Data AutoFilter function can instantly show you where position contains a "2" to see all catchers).

I plan to apply this to several real leagues on an experimental basis this year and will update you on the results later.  I'd also appreciate hearing from any users who have the courage to try it in their own league, perhaps because they always feel they get burned by a single projection or two that doesn't turn out well.  While it's early in the development of the system, if you've tried everything else and failed and want to try something I've used successfully in a very tough 20-team league once, then please do let me know how you make out.

The Skip to the Safe Draft Strategy

(1) We take our ranking chart, in order of descending projected value, and create three extra columns: We label them TAKEN, SKIP and CANDIDATE.

(2) The TAKEN category is simply one that will contain an X to remind us that a player has already been taken in the draft. Having it in a column like this allows us to avoid deleting rows.  We can just look for the first blank space in this column to find the highest-ranked remaining available player.  In non-keeper leagues, this would start as a completely empty column.

(3) The SKIP category is the key to the strategy. We look at the player's projection (not the value but the actual projected statistics). We compare this projection to what the player did last season. This is the part that's going to surprise readers as it's counterintuitive. If the projection is moderately to significantly better than what the player did last year, we put an X in the SKIP column, signifying that we're going to bypass this player in the draft. Yes, you read that correctly. It means that any projected breakout season, substantial improvement year or rookie who's suddenly going to get playing time is going to get an X in this column and that means that we will "skip" the player in question and move on to the next name in the list.  The system has the effect of deliberately conceding ground in that we will not get the benefit of the most surprising, ultimately successful forecasts.  The trade-off is that we will be playing it "safe" here, hence part of the system's name.  We'll let some other fantasy team have that nice projected value, the riskier value that we have decided to skip.

(4) If the player's projected stats are approximately equal or worse (yes, worse) than what he did last season, we put an X in the CANDIDATE column. This means that the player is a candidate for us to draft. As I say, I know it is counterintuitive to tag players who are projected to be worse than last year but what we're doing here, in combination with previous year statistics, is looking for "worst case scenario" players whose projection is worse than a year ago and yet still somehow manage to come out high on our projected rankings.

(5) When our pick comes up, here's how we decide whether to take the highest-ranked CANDIDATE - You cannot take a player in a round if you believe that he's likely to last until your next pick. If you have doubts about whether he'll last, the tie-breaker here is whether you anticipate that this pick will be so shocking to at least a few of the opponents as to be perceived as a risky one.  In other words, doing surprising or shocking things under this strategy is absolutely prohibited.

So, if the player is likely to last until the next round, you skip the player in question, aware that you may miss out on him but in most cases, you will not. The purpose here is to avoid picking up players with an early pick whom you can get later and who have perceived value below what this round should be offering. This strategy is pursuing safe picks with maximized perceived value who also are projected to be valuable to our team if we don't trade them.

(6) Where a fantasy GM's knowledge and skill comes in is in determining whether the player deserves a "skip" or a "candidate" flag because some aren't easy. For example, Daisuke Matsuzaka was an easy skip leading into this draft because we didn't have any 2006 statistics and thus, his 2007 projected value was way higher than his 2006 actual value. But sometimes you're dealing with a player projected to improve 5 or 10 home runs and 20 points in batting average.  Those ones can be tough to flag appropriately and you have to use your skill here to weigh all the categories appropriately.

(7) A "skip" can actually become a "candidate" later in the draft  if the player's previous season (i.e. 2006 if you're drafting for 2007), while being worse than his projection, is still substantially better than the best remaining player who is currently flagged as a CANDIDATE. In other words, there can come a point where the gap in projected value between an available SKIP player and an available CANDIDATE is so wide that I have to go after the best SKIP player.

(8) Perception needs to be considered in a way to discern whether opponents are going for starting pitchers, closers, other relievers, good hitters or decent hitters that seem to be drafted because of position scarcity. Staying tuned and having a sense of the latest trend in the draft will help us decide whether certain players we are thinking about taking might last until a later round.

Let me show you how this was actually applied last night in preparation for the start of the draft. Here's an actual screen capture of the top portion of my strategy sheet from Excel that I described earlier.  The left side is a straight copy and paste from our fantasy domination ranking sheets for a 12-team mixed league with 20 games played position rule.  I've added the three columns at right as per the SSDS approach above:

You can ignore the "taken" column because as I screen captured this at the end of the draft, all the players are now listed as taken.  At the beginning of the draft, this column was empty.  I know some readers will have a hard time grasping the strategy and that's why in the next part of the series, I'll take you through the application step-by-step as it was actually used in this mock draft.

Tomorrow, we apply this approach to "The Draft" and then on Friday, we look at "The Results"