How Do You Want To Lose?

Nobody wants to lose – nobody plans to lose – but if you were going to lose, how would you want it to happen? Would you rather fall behind early and never manage to catch up? Or would you prefer to hold the lead for most of the game only to give it up at the end?

The sooner you fall behind, the more time you have to catch up.

Having lost our first game on a last-second 4th-and-10 defensive break-down, I'm pretty sure that’s the worst way to lose. At least if you get soundly defeated, you can console yourself with the knowledge that the other team was just plain better. But if you lose right at the end, it's not because they were better, it's because they beat you. You got beaten. You let them beat you. If you get an early lead, the best that can happen is you completely crush the other team, which is a hollow victory. If you don't completely outclass the other team but still hold on for the win, well, that's good. But the worst that can happen is that you blow the lead and lose, and that's the worst thing that can happen.

On the other hand, if you give the other team the lead early on, then the worst that can happen is you end up losing at the end. But if it’s close, your team will likely be playing hard to the bitter end, which is good. And the best case scenario in this case is a glorious come-from-behind victory. The further you fall behind, the more glorious the comeback! And that's the best thing that can happen. Not only will you have the win, but your kids will have learned to never give up.

Conventional game theory says you should play to avoid the worst case, so all things being equal, you should play to give the lead up early rather than late. This is not just academic theorizing. In a league with minimum play rules, you will be required to play your scrubs for a certain number of plays. Most teams start their starters – that's why they're called starters – but game theory would tell you to start your replacements. Let's look at a grid with our choices:

	Play subs early	Play subs late
Other team much better	After giving up some big plays early, the team managed to slow the other team's momentum.	The team struggled early on, but then seemed to lose hope as they opened the floodgates to a rout.
Other team slightly better	After struggling early, the team put together a comeback that just fell short.	An even match in the beginning, the team blew it in the end.
Other team slightly worse	After struggling early, the team never gave up, and put together a glorious comeback for the ages.	The team led early on, but let the other team chip away at its lead. After the other team threatened a comeback, our team held on for the win.
Other team much worse	The team took its early lead and only built on its own momentum.	The team took its early lead, and although its momentum slowed, the outcome was never in doubt.

In every case except possibly the last one (when you completely dominate the opposition) playing your subs early presents a better outcome. This tells me I should start my subs – the worst go first – putting the better players, who are more capable at mounting a comeback or preserving a lead, on the field at the end.

What if when you play your subs affect the outcome?

My assumption so far has been that your timing of your substitutions will not affect the outcome; they will just affect the “story” of the game. What if this is not true?

Well, certainly from an operational point of view, it may be disruptive to pull players on and off the field after only one play. The players are going to wear themselves out running to and from the sidelines, and they will be unable to learn from and build upon their experiences from the previous play. I like to keep offensive subs in for a whole drive, which is self-limiting of the poor combinations: if a personnel package isn't working, the drive will be short and the subs will come off quickly; if the package is doing well, then they get to play more as the offense drives down the field. On defense, the opposite is true, but I still wouldn't want to pull a player out after just one play. I'd say leave him in at least three or four.

But this post is more about the strategy of the timing of your substitutions, particularly with respect to the play of your MPP’s (minimum play players). Earlier I had advocated getting their play time over with early in the game so that your more capable players could take the field to clean up the mess. But if you're using your MPP’s to give your two-way starters a break when they get tired, then playing them before your two-ways even get going does not serve this purpose. However, I still think it is important to get all your players their minimum plays in the first half if possible, or by early in the second half, rather than the current common scheme of trying to put multiple MPP’s on the field at the end of the game because they have not yet met their quota.