On Monday night, the Los Angeles Dodgers hosted the Atlanta Braves in Game 4 of their Division Series. The Dodgers were one win away from advancing; the Braves one loss away from elimination. And when the two teams took the field on Los Angeles, the two starting pitchers were Clayton Kershaw -- winner of the NL ERA title three years running and safe pick for best pitcher in the world -- and Freddy Garcia, a 36-year-old, 13-year veteran whose best days are far behind him.

Forget for a moment the outcome of the game, a 4-3 Dodgers victory. Though Garcia gave up two home runs early to Carl Crawford, he had his best stuff and got out of trouble when it came his way. It's true that Yasiel Puig missed a two-run homer off Garcia by about five feet, and a whole lot of well-hit liners and one-hoppers were lucky to find guys in the Braves infield not named Andrelton Simmons, but that's just how it goes sometimes. The interesting bit to examine is the process: Should a team shorten its starting rotation in the playoffs, when, and why?

Monday night was a clear-cut example where both teams had the option to shorten up their rotation, if they believed it put them in the best possible position to win. The Dodgers did it, moving Ricky Nolasco into a reserve role to start their ace Kershaw on three days of rest, in effect going all-in on winning Game 4 at home, which would secure an extra rest day and head off a cross-country trip back to Atlanta for Game 5. The Braves had the option to shorten up as well; Kris Medlen, the Atlanta starter who faced Kershaw in Game 1, also could have pitched on three days' rest. Medlen, who spent a long stretch of 2012 as a swingman/long-relief type, before moving into the rotation for an impressive streak of starts to end the season, is no stranger to multi-inning outings on short rest. Unless there were injury reasons to keep him on the bench -- and we've not yet heard of any -- he'd have been a far stronger play than Freddy Garcia on full rest. Was he ideal? No. Did the Braves almost get by without him? Yes, because that's how baseball works sometimes. Does making a silly managerial choice once and almost getting by mean that it was really the smart move all along? Ask Craig Kimbrel about that one.

Both of the Braves' big pitching decisions Monday night -- starting Garcia, and limiting Kimbrel to a four-out save that never came -- were conservative managerial decisions. They're regular season managerial decisions, born out Fredi Gonzalez's apparent unwillingness to recognize that the stakes change in the playoffs, and the calculus on future returns from your players changes with them. No one particularly cares if Kimbrel would be tired today or Wednesday or Thursday, because the Braves aren't playing anymore. Similarly, as long as the staff watches Medlen closely, the chances of a guy with a history of pitching in relief suffering a random, catastrophic injury in a game on his regular throw day aren't that much higher than at any other particular time he's throwing at game speed.

We have no real direct, hard correlation, let alone causation, between raw innings pitched totals, raw pitch counts in individual games, and arm injuries; what we do have is the recognition that the more you pitch, the more likely you are to get tired, and when a pitcher gets tired but still throws at game speed and his manager doesn't pull him, that's when injury rates climb. And then, of course, sometimes guys just get hurt. That's part of the game. But there's a difference, especially in the playoffs, between managing pitchers intelligently and managing them out of fear they'll get hurt. For the record, I'm not convinced any of this had much to do with why Gonzalez started Garcia; I just think it should have.

Managing pitching intelligently is precisely what Mattingly and the Dodgers did by starting Kershaw against Garcia, however. It too was conservative, but in another way: "in line with how baseball used to be played." The focus of the worry was Kershaw's 125 pitches in Game 1, which was misplaced though well intentioned. There are pitchers for whom throwing 125 pitches is a problem, who visibly flag under that kind of workload late in the year, especially after 200-plus innings pitched. Those pitchers were not Clayton Kershaw in Game 1 of the NLDS. The Dodgers ace was throwing free and easy all night in the first game of the series, save some command problems in the middle innings, and clearly that gave the Dodgers confidence that he could give LA something on three days of rest.

But the number 125 jumps out at people, causing cold sweats and Mark Prior jokes. It makes people say silly things, like that they'd prefer to start Ricky Nolasco instead of Clayton Kershaw, because the Dodgers could afford to lose Game 4. Pitcher injuries can be scary, but when we're so afraid of them that we'd argue for suboptimal matchups out of fear (and Ricky Nolasco in front of the Dodgers defense that showed up Monday night probably ends very badly), then we've lost perspective. The question is not: "If he doesn't start, why is Ricky Nolasco on the roster?" Putting a guy on a playoff roster is not a suicide pact; managers should put whichever players out there will give them the best chance to win.

The Dodgers have done just that. Kershaw is now in position to start Game 2 of the NLCS on normal rest, with Greinke getting the ball in Game 1, positioning both players to bookend the series for LA with two starts each -- or, if the going gets tough in the middle of the series, to make further appearances on short rest, so long as they're not showing any signs of arm fatigue. Pitchers like Clayton Kershaw are extremely potent weapons to have in the playoffs, and Don Mattingly has shown so far that he knows how to use them.