OT is fine ...

just not the current overtime framework associated with college football. For now, I'll focus on the option of going for a two-point conversion before it is required. Allowing it is not equitable ... thus, it should not be allowed.

Rewind back before the OU v. Boise State game to the Oregon State v. Missouri match-up. Oregon State (with help from the officials) was able to pull to within one point late in the game (35-34). They could have kicked the extra point and tied it. Instead, they went for two. They converted and won the game. That was not fair to the Tigers. You should not be allowed to go for two to take the lead late in the game because it comes down to who scores when. You shouldn't be penalized for scoring early in the game, should you? All congruent scores should count the same.

The situation came up when Boise State went for two to beat Oklahoma. Some might argue that OU could have gone for two in a preemptive strike. That is true, but Bob Stoops would have been taking a major risk if he'd gone for two after the Peterson TD run on the first play of OT. Stick with me and you'll soon understand why it matters who has the ball first and why it isn't fair for the team that has the ball second to make a choice.

To keep it neutral, we'll use A and B to represent the two teams. Keeping it simple, A will have the ball first and B will have the ball second. Also, for simplicity, we'll assume the probability of scoring on a 2-pt attempt is 50% and we'll assume that kicking the extra point is automatic, neither of these is a huge stretch. So, each team can either (1) kick the extra point, (2) convert a 2-point try, or (3) fail on a 2-point try. Obviously, if A converts a 2-pt attempt, B will not kick the extra point. Also, if A fails to convert, B will kick the extra point rather than risk going for two.

So, we have the following permutations.
A kicks : B kicks -> Tie
A kicks : B fails -> A wins
A kicks : B converts -> B wins
A fails : B kicks -> B wins
A converts : B fails -> A wins
A converts : B converts -> Tie

Wait, there are 6 combinations. Yep. Two result in a tie, two result in A winning and two result in B winning. Exactly. What's not fair? The important point is not how many different permutations there are, but the probability of each happening. Follow?

If A kicks, B can tie by kicking. If A kicks and B goes for two, it's 50/50 on who will win, theoretically. But, A does not have the same opportunity as B. If A goes for two, half the time they will convert and half the time they will not. That's simple. So, when A does not convert, B can simply win by kicking. Thus, B wins half the time A goes for two. But, because B has the option of going for two if A converts, B will tie the game half of the 50% of the time A converts. So, A will win 1/4, B will win 1/2 and a tie will occur 1/4. Those odds are definitely with Boise State if Stoops had gone for two. Wait, I mean the odds are for B.

A simple fix to make it equitable, and still allow both teams the opportunity to take the risk of winning/losing would be to allow A a shot at going for two to tie to prolong the game if B attempts and converts a two-point play following A kicking the extra point. So, using the OU v. Boise State game as an example, OU scored and kicked the extra point to go up 42-35. Boise State scores to answer and make it 42-41 OU. BSU goes for two and goes up 43-42. OU would have a chance to counter the BSU two-pointer and retake the lead 44-43. BSU would then have to convert the extra point to make it 44-44 and the 2nd OT would begin. That would even the odds.

A kicks. B goes for two.
Half the time, B loses.
Half the time, B converts and goes ahead, forcing A to try to counter.
Half of the half, A converts and forces B to tie ... resulting in a tie and subsequent OT period.
The other half of the half, A fails and B wins.
Thus, B wins 1/4, A wins 1/2, and it's a tie 1/4.

Wait, that's not fair ... then B is at a disadvantage! Wrong! It is fair because the team that initiates the 2-pt conversion attempts assumes the risk and it does not matter which team goes first as far as the statistics are concerned, which is not the case currently.

If both teams are forced to go for two, that is fine too.
A fails : B fails -> Tie
A fails : B converts -> B wins
A converts : B fails -> A wins
A converts : B converts -> Tie
That part of the system is fair for both teams, at least as far as going for two goes.

Check back periodically for more discussion of flaws in the college football OT system.