The Fallacy of 5 Starters

There has been a lot of discussion lately about trading Gio. And many of those suggesting it do qualify that suggestion by saying that the Nats should first sign a Free Agent starter.

In 2015, the Nationals had 9 different pitchers who started games:

  • Max, Jordan and Gio all started 30+ games.
  • Strasburg started 20+ games
  • Roark, Ross and Fister started between 10 and 19 games
  • And two pitchers, AJ Cole and Taylor Jordan each started one game.

Another way to look at this is that of the 5 projected starters in 2015, the 4 with the most starts, started 120 of their 162 games. That left 42 games to be started by others. And the Nationals were not atypical compared to the rest of MLB. The following table shows how many starters each MLB team had (starters were determined based on who threw the first pitch in each half of the 1st inning). For example, the Angels (ana) had nine pitchers who started a game; 2 of them started 30 or more games; 3 started between 20 and 29 games; 1 started between 10 and 19 games; 1 started between 5 and 9 games; and 2 started 4 or fewer games. And note that 21 teams had 10 or more different pitchers that started a game. In fact, the Nats were among the teams that had the fewest number of different starters.

# of Starters by # of Starts


TEAM All 1-4 5-9 10-19 20-29 30+
ana 9 2 1 1 3 2
ari 12 3 3 2 3 1
atl 11 2 3 3 1 2
bal 8 . 2 2 1 3
bos 12 4 2 2 3 1
cha 9 2 2 . 1 4
chn 10 3 2 1 . 4
cin 14 4 3 4 2 1
cle 11 4 1 2 . 4
col 12 2 2 4 4 .
det 12 1 4 3 3 1
hou 13 4 3 3 1 2
kca 10 2 2 2 3 1
lan 16 9 1 2 1 3
mia 13 3 1 7 1 1
mil 11 2 3 . 5 1
min 9 1 1 3 2 2
nya 10 3 . 3 4 .
nyn 10 3 2 . 3 2
oak 13 3 4 3 2 1
phi 14 2 5 4 3 .
pit 8 1 1 1 2 3
sdn 10 3 2 1 . 4
sea 10 2 . 3 4 1
sfn 9 1 1 3 2 2
sln 9 2 2 . 2 3
tba 9 1 1 3 3 1
tex 12 3 1 5 1 2
tor 12 5 1 2 2 2
was 9 2 . 3 1 3

So if one of our starters is traded (Gio being the most frequently suggested), do you really want to count on any of these pitchers for too many games (as stated above, the top 4 starters in 2015 started 120 games; and the Nats top 5 starters started 135 games) in 2016:

  • AJ Cole
  • Taylor Jordan
  • Taylor Hill
  • Lucas Giolito (and thinking he is going to be ready is a real stretch)

The Nats need to sign another starter (and maybe 2) for 2016. You need to have at least 6 MLB capable starters. Both Ross and Roark have options left – so one of them can be stashed in Syracuse if need be. Ross was added to the 40 man roster in 2015 and was not optioned back to the minors, so he has all his options left. And Roark was added to the 40 man roster in 2013 and was sent down to minors for 10/11 days in 2015. However according to FanGraphs Player Options page, any player who spends less than 20 days in the minors does not use an option. Depending on whether that is still the case, Roark would have either 2 or 3 options left.

Looking at the MLBTR Free Agent Tracker, none of the remaining starting pitchers on the market are terribly exciting. Chen, Gallardo and Kennedy have QOs, so signing one of them would involve losing the first sandwich pick. And it is not clear than any of them would be as good, let alone an improvement, over Gio. So when evaluating a trade involving Gio (and others) for someone like Blackmon, consider that you are replacing Gio with a more expensive pitcher who may not be as effective, and you are giving up a pick.

Bottom line is that trading a starter may not be a good idea.

This entry was posted in Analysis, FreeAgents, Gio, Sabermetrics, Starters. Bookmark the permalink.