As I discussed in my article on how to mock draft like a pro, it's certainly possible to do more than just "pass time" while mock drafting. You can also learn something along the way. For example, what might I discover by partaking in multiple drafts, each time selecting from a different starting spot?
There's a practical purpose to this exercise, as it is becoming increasingly popular to allow draft participants to choose their draft spot. Alternatively, some leagues reveal draft positions only very close to the draft date, rendering it difficult to partake in many practice drafts from your assigned spot. Having experience drafting from everywhere can help you out, regardless of what pick you've got.
Now, while it's advantageous to simply draft from every position, even more can be gleaned by turning the draft room into a laboratory and designing a specific experiment. Of course, the best strategy is to always keep an open mind and take what the room gives you. However, almost everyone has a preferred approach. That said, regardless of which spot you're assigned, leaving your draft with the best-case scenario result requires that you choose the optimal first pick when given the opportunity.
To this end, I organized a Mock Draft Decathlon, setting up 10 separate mocks and inviting guests to join me for some spirited drafting over the course of a week. More than 60 individuals took part, with several doubling and even tripling up. I started from a different spot in each leg of the decathlon. Each draft assumed a 10-team league with standard 5x5 rotisserie scoring. Rosters consisted of 1-C, 1-1B, 1-2B, 1-3B, 1-SS, 1-CI, 1-MI, 5-OF, 1-UT, 9-P and three reserves.
While the act of enforcing restrictions violates one of the main tenets of drafting (go with the flow), in order to test the variable of initial draft position, everything else I did had to be controlled as much as possible. The following rules were strictly followed throughout the Decathlon:
Rule No. 1: I must select J.T. Realmuto once he's the highest-ranked player on my draft list
There is a proper way to rank catchers and, quite frankly, there's a significant portion of the fantasy community doing it incorrectly. This offers up a huge buying opportunity for those doing it right. Each position is its own pool and the worst draft-worthy catcher produces stats of lesser quality than the worst draft-worthy player at every other batting position. Hence, a pricing adjustment must be made.
Consider a two-team, four-player league scoring just homers. Each team needs to draft a catcher and an outfielder. Here is the player pool: C1, 20 projected HRs. C2, 10. OF 1, 40. OF2, 35. You have the first pick. Who do you want?
Hopefully, you chose C1, leaving me OF1 and C2, for a total of 50 homers. When you add OF2, your sum is 55. The key here is that only the unique, or useful, homers matter. We both will have at least 10 homers from our catcher and 35 from our outfielder. However, C1 contributes 10 more useful homers than C2, while OF1 produces only five more useful homers than OF2.
While this is a bit of an oversimplification, if you apply this notion to ranking the full fantasy catcher pool, there are advantages to be gained when drafting backstops as the unadjusted rank is much lower. In most years, this benefit extends to a handful of receivers, thus the trick is discerning the optimal time to pounce. However, according to my projections and rankings, the delta between Realmuto and the field -- even in a 10-team, one-catcher league -- is so steep, it's either take Realmuto or wait until the end.
Rule No. 2: I must draft one ace from among Jacob deGrom, Gerrit Cole, Max Scherzer, Justin Verlander, Walker Buehler and Jack Flaherty
The 2019 National Fantasy Baseball Championship Main Event competition consisted of 38 15-team leagues, using standard 5x5 scoring but with no trading. The 570 teams were lumped together to crown a champion. Of the top-100 teams, 67 of them selected a pitcher by the end of the second round. Even though this is a different format, there was ample evidence to force an ace onto my mock teams.
Rule No. 3: I must draft a top closer from among Josh Hader, Kirby Yates, Roberto Osuna and Liam Hendriks
Admittedly, especially when drafting a top starter, I tend to wait on closers. However, I thought this would be an intriguing condition to impose since it involves "reading the room" to make sure this requirement gets fulfilled.
Rule No. 4: I must incorporate ample multiple-position eligibility
My goal here is that each infield spot (1B, 2B, SS and 3B) has at least two eligible players, and that I leave the draft with a minimum of seven OF-eligible players. Players eligible at more than two positions count at each spot. Keeping in mind that my ultimate goal is "winning the league" and not "winning the draft," it's beneficial to sacrifice raw stats in order to bake in a flexibility chain. Your goal should be having the ability to activate the best reserve or to pick up the best player on waivers and fit him onto your roster, regardless of position. In other words, you're not drafting a player. Rather, you're drafting a roster spot with the selected player being the first to populate it. Over the course of the season, several spots will be occupied by different players each week. The production you might lose by leaving a "better" player on the table is made up by the flexibility to add more in-season production across your entire roster.
Rule No. 5: I must avoid stolen-base specialists unless absolutely necessary
To be honest, in 10-team leagues, this is a no-brainer. In the current inventory, only Mallex Smith qualifies as a draft-worthy SB specialist. In deeper leagues, other individuals come into play (Jarrod Dyson, Delino DeShields), but the days of deciding whether or not to draft Billy Hamilton or Dee Gordon early are history. Still, with so many other constraints, there's a chance the roster would be devoid of speed, so Smith could come into play.
Rule No. 6: I must not refer to the real-time standings or track category totals
This is how I usually draft, so it made sense to impose this rule in this exercise. Understanding target drafting is common. However, I'm not a fan of this process. Again, my motto is, "Win the league, not the draft." Too many fantasy managers draft with a goal of finishing at the top of the real-time standings. With a season's worth of roster management ahead of you, the better approach is to collect the strongest foundation. Real-time standings don't account for streaming pitchers or replacing injured players. My research shows that anywhere from 65% to 90% of season-ending stats are acquired via the draft, with a shallow 10-team league falling on the lower end of that scale. I'm not interested in finishing with the most roto-points at the end of the draft. I want the highest total at the end of the season. Not to mention, there's built-in bias using one's own projections to fuel the standings. Since I'm the point person for the ESPN fantasy baseball projections, I better finish at (or near) the top of the draft standings. That said, I'm likely to end up much lower when a different set of projections are used.
So, let's take a look at the 10 teams I drafted, and then we'll review how well I followed my rules.
Rule No. 1: Realmuto ended up on eight of my 10 rosters. His ADP was 36, or mid-Round 4. I grabbed him in the fourth round seven times and one time in the fifth. Though, to be fair, the fifth-round pick was at the turn. When I had the No. 4 pick, Realmuto was selected 41st overall. With the 37th pick, I chose Jose Altuve, ranked just one spot higher than the backstop. When I had the No. 10 pick, Realmuto went 46th. I had again drafted Altuve on my previous turn, 31st overall. Had Altuve been off the board, I would have been "forced" to take Realmuto at that spot -- which would have been the earliest of any of the mocks.
One of the limitations of this restriction is not having an accurate market read on Realmuto. In a real draft, the plan would be to split the difference between where I have him ranked and where the market would take him. After the mocks were complete, I polled all the participants as to when they would have drafted J.T. Realmuto. The results pointed toward a late-Round 8, early-Round 9 ADP. That means targeting Realmuto when I did was the right timing to assure rostering the Phillies bell cow.
Of course, this doesn't address whether or not it was most beneficial to guarantee Realmuto was on my rosters. On paper, forcing a fourth-round pick weakens a lineup that could have snagged him later. However, the real question is how a team forcing an early Realmuto compares to one waiting on a catcher, since forcing the backstop minimizes (if not mitigates) the return on investment.
Rule No. 2: Mission accomplished, although this was helped a lot due to the 10-team format. In a 15-team league, ADP suggests that the only way to grab an "ace" with the top two draft positions would be to take them in the first round, since the latest any of my designated aces was drafted was 28th overall (Verlander). It's interesting to note that, with the No. 2 pick, I was able to wait until Round 3 and still roster Buehler. When I was up in the second, Verlander and Flaherty were both also still available, so even if aces were double-tapped at the wheel, a third would still be on the board. Scherzer was my most popular ace, appearing on half the squads. It could have been timing, or perhaps a reflection of my confidence that he'll be raring to go once the season commences, while others may be reticent about last season's injury woes as well as a couple of minor issues in the spring.
Rule No. 3: Once again, we did it! The plan was to wait until Hader was picked, then to take a closer on the ensuing turn. It was risky, since drafting the first closer often initiates a run, but fate was on my side -- at least, it was in this exercise. Yates and Hendriks each made it on four of my staffs, though I prefer Yates by a decent margin.
Rule No. 4: To be honest, this was easier than expected. Although, in retrospect, I should have known this to be the case as there are more players currently eligible at multiple positions than ever before. It helped that I was well ahead of the market on Eduardo Escobar, as well as with J.D. Davis and Starlin Castro. Full disclosure, I counted Jean Segura as a multi-position player since he'll quickly add third base to his shortstop qualification.
Rule No. 5: Mallex Smith ended up on two of my rosters. On one, he was taken in Round 20 and placed at utility. In the other, the speedster was picked in Round 23, as my first reserve.
Rule No. 6: You're just going to have to trust me on this, but I refrained from peeking. As I previously explained, since I head the team responsible for the site projections, there isn't much to be learned by these standings, since I should organically be at the top. A salient argument can be made that it's still important to track the balance between homers and steals. Humbly speaking, though, once you do this for as long as I have, that becomes intuitive.
While judging my results versus each individual league isn't likely to be very fruitful, pitting my 10 lineups against each other could elucidate a lot. Again, all we're looking at is a foundation, but ranking my results roto-style speaks toward which one offers the best foothold to embark on the season.
The primary objective of the experiment was to try to determine the best draft position to carry out the defined strategy. While it's not definitive, favoring an end pick is best, followed by the middle, with an early pick looking to be the least beneficial. This is surprising on a couple of levels. Firstly, I personally prefer a wheel pick, starting at either 1, 2, 9 or 10. Three of the poorest squads come from this group. However, the imposed restraints could have something to do with this discrepancy, as well as the reason I like picks at or near the turn is that they facilitate roster construction -- and some of that freedom was voided by the rules.
The second reason has to do with the way rankings manifest when dealing with projections. The delta between consecutive players is greatest at the top of the pool. As you serpentine through the field, those differences diminish. However, the early gaps are so large, snaking doesn't even things out. If you assign the corresponding auction price equivalent to each draft spot, the early teams essentially begin with more than the standard $260, the middle teams have a dollar or two less while the end teams are at $260 or maybe a dollar more. The season-ending stats will obviously differ greatly from the projections, so the "on-paper" benefit rarely comes to fruition. Still, in an exercise of this nature, the bonus production of the first few picks -- while accounted for in the standings -- is not reflected in the order of finish.
There are two striking observations even more germane than starting point. The two best teams accrued a lot more pitching points, while dumping a hitting category. Neither was planned, but perhaps it is serendipitous. The top team finished last in batting average. This may seem like a difficult hurdle to overcome, but in fact it isn't a big deal. It's hard to see just by looking at category rankings, but batting average is the most bunched category. That is, if the categories were normalized to contain the same number of units, the top-to-bottom spread in average is the smallest. Plus, batting average is the most volatile of all categories. Projections are treated as static, but if error bars were associated with each category, the high end of the batting-average range would overlap the low end of several teams, suggesting that luck itself could propel a team way up the category ranking.
Perhaps more importantly, there's a lot of room for improvement in the pitching ratios, which is a primary target of in-season management. There's also upward movement available in strikeouts -- another benefit of streaming. It may seem like the team lacking in steals is in better shape than the offense with the lowest batting average since shuttling in speedsters over the course of a season could supplement the deficient category. On a game-theory basis, this has merit since the SB category is generally more bunched in the middle than the other batting counting stats. That is, it doesn't take much to move up in the category. The question is, by doing so, are points lost in other places? Plus, this particular team funneled assets not used for steals to improve pitching ratios, minimizing the extent at which ERA and WHIP can be upgraded, although strikeouts and wins can be improved.
Obviously, analysis can only be taken so far since every draft is its own entity. Observations, let alone conclusions, may not transcend the real thing. However, looking at the roster construction of the various clubs in concert with the preset restrictions avails several questions with answers aiding in plotting a strategy: How important is it to draft J.T. Realmuto? Is an ace really necessary? Is a top closer really necessary? Should the foundation be more hitting- or pitching-heavy? How much does having freedom to make any pick aid roster construction?
Sure, setting up more Decathlons could answer these questions, but fortunately it isn't necessary. All the required data is already available from the overall ADP. (Well, almost, as the ADP of players like Realmuto, Charlie Blackmon and Mike Minor are skewed since my picks, and not the market, set the value.) Even so, shadow drafting the ADP from different draft spots should lend some insight to the above questions. The best part is that this is something you can do as well, though finding a suitable ADP could be an issue. That aside, a follow-up to this discussion is in the works, detailing how to shadow draft along with presenting the resulting teams with analysis and how shadow drafting can help mold your strategy.