GM's OFFICE: Observations from Memorial Day ADPs

Our friends at the National Fantasy Baseball Championship put together a "second chance" contest, with a series of drafts held over Memorial Day weekend and playing out over the rest of the season. These are 12-team mixed 5x5 (BA, not OBP) with both league and overall prize pools.

The NFBC folks were kind enough to post the ADPs from the event (here, choose 'in-season' from the second dropdown list). I didn't enter this contest, but I spent some time this week looking through these ADPs, and found them fascinating.

The double edges of ADP

As we talk about a lot during the offseason, ADPs are both a blessing and a curse. But in this specific implementation, I find that the plusses greatly outweigh the minuses, for two reasons:

1. One of my big criticisms of the preseason NFBC ADPs is the echo chamber effect: the ADPs start populating after the first drafts, before the calendar year even rolls over. Everyone starts looking at those small-sample results and referencing them in their own drafts, and those initial ADPs crystallize and become gospel by March. But in this contest, all of the drafts (15 of them) were held within less than a week start to finish. That's a nice sample size, and concentrated enough to avoid that ADP-induced groupthink effect.

2. In the preseason, every drafter is working from their own projections, whether created themselves, or from their projection system of choice. That's still the case here, of course, but there's another piece of important information that everyone has equal access to—the current year-to-date stats.

Thanks to these factors, these drafts seem like more of a pure valuation and evaluation exercise; there's more of a level playing field of available information, and less groupthink.


In all of these following charts, I'm going to show three rankings:

MayADP: The May ADP from last week's drafts
MarchADP: The preseason March ADP (taken only from NFBC Main Event drafts held in mid-late March, so as not to muddy that ADP with risers/fallers from those early-winter ADPs)
YTD Rank: Year-to-date mixed 5x5 value, as taken from the Custom Draft Guide, but converted to a 1-to-x ranking, so it looks more like an ADP of sorts.

The top of the draft board

Let's start at the top and look at how the first two rounds of these drafts played out:

  Player May ADP March ADP YTD Rank
  Trout, Mike 1 1 33
  Yelich, Christian 2 5 3
  Bellinger, Cody 3 40 1
  Betts, Mookie 4 2 31
  Arenado, Nolan 5 10 6
  Acuna Jr., Ronald 6 6 36
  Lindor, Francisco 7 20 162
  Verlander, Justin 8 14 2
  Turner, Trea 9 4 386
  Mondesi, Adalberto 10 36 8
  Story, Trevor 11 13 10
  Bregman, Alex 12 16 28
  Martinez, J.D. 13 7 56
  Scherzer, Max 14 3 116
  Baez, Javier 15 22 18
  Sale, Chris 16 11 200
  deGrom, Jacob 17 9 174
  Freeman, Freddie 18 25 24
  Cole, Gerrit 19 17 53
  Merrifield, Whit 20 32 25
  Snell, Blake 21 28 92
  Sanchez, Gary 22 54 101
  Bryant, Kris 23 37 37
  Ramirez, Jose 24 8 233

Lots to chew on here:

We have long known that our hit rate on preseason first round picks is astonishingly low, around 35%. At least in this expanded view of 24 picks, though, the turnover isn't that bad: 17 of these picks came from within the preseason Top 25.

But wait! Only 10 of these 24 have actually returned Top 24 value through the first two months. So even with that in-season data (and a decent sample size of it!) available, there is still a fair bit of projection going on here, rather than just expecting early performances to continue to hold up. That might be savvy analysis, or it might be over-reliance on preseason expectations and/or being too slow to adapt to what's happening in front of our eyes every night.

Specifically, after the second-half fade last year and now 200+ ABs of struggle this year, it takes some serious fortitude to take Jose Ramirez at the back of the second round in these drafts. I don't think I could pull the trigger on that... as a point of reference, our 12-team mixed projection for Ramirez sits at $21 for the rest of the year (16-51-.264-22 SB-55 Runs), which certainly isn't bad, but isn't worthy of pick 24 either.

Risers and fallers

Next, here are the biggest ADP risers from March to May (filtering on the Top 120 picks/10 rounds in these May drafts):

  Player May ADP March ADP Mar->May delta
  Riley, Austin 79 n/a n/a
  Soroka, Mike 84 n/a n/a
  Chavis, Michael 99 n/a n/a
  Neris, Hector 115 n/a n/a
  Smith, Caleb 86 329 243
  Bell, Josh 31 224 193
  Boyd, Matthew 119 296 177
  German, Domingo 114 286 172
  Tatis Jr., Fernando 104 254 150
  Alonso, Pete 65 204 139
  Senzel, Nick 85 198 113
  Reyes, Franmil 105 217 112
  Greene, Shane 110 219 109
  Contreras, Willson 38 135 97
  Meadows, Austin 80 174 94
  Ryu, Hyun-Jin 76 169 93
  Polanco, Jorge 108 196 88
  Paddack, Chris 78 162 84
  Moncada, Yoan 61 140 79

Youth will be served! Now that there is no waiting period for the rookie class, they're understandably very popular picks relative to their March price. The larger point, though, seems to be a question of roles. Whether rookies or other types of playing time questions (see Franmil Reyes, Austin Meadows), almost everyone on this list had some form of question about their playing time/opportunity back in March. Really, the only exceptions here are Bell, Boyd, Greene, Contreras, Ryu, Polanco, and Moncada. Remember, as always: there's no longer such a thing as having 'no path to playing time'.

This is a fun list: players taken in Top 10 rounds in the May drafts, with YTD value outside the Top 200:

  Player May ADP YTD Rank
  Stanton, Giancarlo 97 892
  Guerrero Jr., Vladimir 30 528
  Turner, Trea 9 386
  Nola, Aaron 40 380
  Hicks, Jordan 89 358
  Judge, Aaron 67 351
  Syndergaard, Noah 59 344
  Riley, Austin 79 322
  Senzel, Nick 85 304
  Clevinger, Mike 111 289
  Wheeler, Zack 113 287
  Carrasco, Carlos 49 262
  Doolittle, Sean 87 259
  Altuve, Jose 44 250
  Treinen, Blake 74 247
  Iglesias, Raisel 95 243
  Ramirez, Jose 24 233
  Paxton, James 81 226
  Tatis Jr., Fernando 104 215
  Sale, Chris 16 200

There are a few buckets making up this list:

1. The rookies who just arrived (Riley, Senzel, Vladito, etc) and have clear opportunity for the rest of the season. Their presence here is quite understandable.

2. The "injured but should be back soon" crowd. Again, the presence of Stanton, Judge, Tatis, Clevinger here is understandable. But putting these guys in context of a March draft, you could have said the same about Luis Severino or Justin Upton back in March, and their owners are still waiting on them, two months later. Even those who took Altuve above are sweating now compared to when they drafted him just a week ago. Even James Paxton and Trea Turner, just off the DL, still carry some risk of setback or just lingering underperformance.

3. The "I don't know what's going on, but this seems like a good spot to take him" group. There's Jose Ramirez again, or his pitching doppleganger in Noah Syndergaard. Their ADPs are essentially season-riding-on-being-right bets on these guys. I'll place a wager that nobody who has had to stomach owning these guys for the first two months were doubling-down on them in these drafts.

Spinning the above chart on its axis, here are the members of the YTD Top 75 earners whose May ADP fell outside the Top 100:

  Player May ADP YTD Rank
  Odorizzi, Jake 170 15
  Boyd, Matthew 119 20
  Giolito, Lucas 158 22
  Glasnow, Tyler 176 27
  Escobar, Eduardo 129 29
  German, Domingo 114 35
  Marte, Ketel 126 38
  Polanco, Jorge 108 40
  Santana, Domingo 153 41
  Andrus, Elvis 122 42
  Woodruff, Brandon 143 43
  Minor, Mike 154 46
  Chirinos, Yonny 270 48
  Kepler, Max 161 51
  Pence, Hunter 237 52
  Dozier, Hunter 171 60
  Gordon, Alex 191 61
  Chapman, Matt 118 64
  Montas, Frankie 134 67
  La Stella, Tommy 174 69
  Voit, Luke 112 72
  Mancini, Trey 138 73
  Swanson, Dansby 234 74
  Hosmer, Eric 194 75

Call this the "nobody believes in us!" list. Putting aside the injury cases (Glasnow), it seems to me that there is a TON of rest-of-season value in this chart, relative to their May ADP. Maybe less so with the pitchers, as issues like innings limits for the kids, or environmental factors (Minor) might temper the upside. But particularly for the hitters on this list, I would be buying nearly every single one of them at their listed ADP.

Bringing it home

For those of us who didn't do one of these drafts last weekend, how do we use this information as we continue to manage our full-season teams? More than just a curiosity, I think there's actionable info here. Take another look at the two tables immediately above—compare the name recognition of the star-studded list two tables up (the "overdrafted based on YTD values" list) to the "nobody believes in us" immediately above. The difference is stunning, isn't it?

Back when he still ran, Ron Shandler used to do a "May Day deal" exercise where he would make a mock trade on May 1 where he would sell a bunch of hot starters for a bunch of slow starters, and the slow starters would inevitably outperform the hot starters over the balance of the season. I tried that exercise in this space once or twice more after we took over for Ron, and it didn't work as well.... the environment seemed to change. Further, and more importantly, it isn't May 1st. It's (nearly) June 1st. The first third of the season is wrapping up. The most egregious luck-based April starts have already corrected (remember Tim Beckham's week of stardom, and the Chris Sale hand-wringing? Those seem like ages ago.)

So, if you own someone on the star-studded list two tables up, these draft results suggest you still have a window to trade out from under them, rather than just being forced to sink or swim with them. And if you want to go trade for an Eric Hosmer, or Tommy LaStella, or Alex Gordon, there's a window to do that.

Back when I was writing the Speculator column, I would do a "buy high, sell low" article around this time each year, theorizing that you could steer into a competitor's instinct to "sell high" on a hot starter, and if you did your due diligence and believed in the early performance, could buy it at what would amount to a discount, all while making your trade partner believe that they were in fact "selling high" on their asset. These ADP results make me think that approach is perhaps more viable today than ever before.

Now, go and work the trade lines!


Click here to subscribe

  For more information about the terms used in this article, see our Glossary Primer.