When Should The Blue Jays Pull Their Starters?


Sometimes a manager just can’t win.  Whatever decision he makes, it is wrong.  If it fails, it was a bad idea to begin with.  And if it does succeed, he should have done it sooner.

One area where this is very evident is the decision to pull a starting pitcher and go to the bullpen.  If the manager waits until the starter falters, he has waited too long.  But if he pulls a starter who is still pitching well, and the bullpen does not deliver, then “clearly” he should have continued with his starter.

Dave Cameron wrote an excellent article on fangraphs describing this phenomenon as it applied to a Mets game in which the starter, Matt Harvey, was pulled for a reliever who promptly lost the game.  In that article, Cameron talks about the “third time through the order” effect, which describes the decline in pitchers’ performance the third (and subsequent) times through the batting order.  In Harvey’s case, the effect was substantial – Matt’s OPS goes from .562 and .567 the first two times through the order to .710 for the third time.  Cameron makes the point that, given this effect, it makes sense for teams to pull starters early, before they show signs of faltering, provided that the team has a quality rested bullpen to take over.

This argument is rarely followed in the regular season, as this would limit starters to ~five innings per game and few bullpens can handle four innings every game.  But in the playoffs, when wins are so important and the bullpen can be pushed a bit harder – what then?  And more specifically to the Blue Jays – which of their pitchers show the “third time through” (TTT) effect the most strongly, and accordingly might merit an early hook?

The Blue Jays pitchers do not appear to fall into the conventional TTT effect, but possibly for different reasons.

More from Toronto Blue Jays News


R.A. Dickey

‘s case, hitters do not learn much from the first two times they see him in a game – he is primarily a knuckleball pitcher, and that is what they will see no matter which time through the rotation.  Dickey’s opponent OPS of .744 the third time through is almost identical to his .726 the first time through.

David Price is another exception.  His .429 OPS the third time through is actually better than his .583 and .567 the first two times through.  Whether this is due to his true “ace” status or to his exceptional control of at-bats, there is no indication of a TTT decline.

Marco Estrada also shows little evidence of a TTT decline.  His TTT OPS of .620 is very similar to his .597 for the first time through, and even better than the .720 he achieved in 2015 on the second time through.

In Marcus Stroman‘s case, we only have a very small sample size and so it is more difficult to give weight to the figures.  But a TTT OPS of .618 does not appear to indicate any major decline.

The bottom line?  While the TTT effect is real, and while we would hope that Gibby and the Jays are aware of it and take it into consideration, it does not appear that any of the starting Jays pitchers have been particularly susceptible to the effect so far in 2015.  So TTT should be a factor, but likely not the primary factor.

Next: Toronto Blue Jays ultimate playoff X-factor: Aaron Sanchez