Don’t fill out your bracket based on a November result

What happens in Vegas, stays in Vegas.

I certainly didn’t come up with that saying, so why am I the only one who seems to remember it? Last week, I posted my full NCAA tournament bracket breakdown one month before the real deal. One half of my bracket featured a Final Four matchup between Creighton and San Diego State. I picked the Bluejays to win and advance to the national title game, and I was flooded with tweets like these:

I understand scanning the schedule when it comes to trying to predict future success. I just don’t agree with it. Not this season. Not for a game in November. Yes, the earlier meeting between Creighton and San Diego State in Las Vegas provides a snapshot of what could happen (and did happen at a certain singular moment in time), but suggesting that me picking against that result is some hot take … I think that’s a little much. Stop me when I say something outlandish:

Denzel Mahoney is now active (+8 points)

The junior transfer from Southeast Missouri State was ineligible to play when Creighton and San Diego State met in November as he satisfied the NCAA’s transfer rules, but he’s in the mix now for the Bluejays and that makes a big difference as he has poured in at least 14 points in the majority of games this season. Of course, it’d be irresponsible of me to simply add those 14 points to the bottom line when looking back at this November game, so follow me here.

In the first meeting with SDSU, Shereef Mitchell and Jett Canfield (both of whom have seen their minutes cut since Mahoney gained eligibility) combined for 10 points in 37 minutes of action. In recent action, Mahoney is trending toward 30 minutes and Mitchell is more in the 6-10 range. If we assume that the 37 minutes from Mitchell/Canfield in Game 1 are split 30 for Mahoney and seven for Mitchell in this fictitious Final Four matchup and that both players score at their current per-minute average, the expectation would be to get 18 points, or eight more than this role generated on Nov. 28.

Ball security (+8 points)

The Bluejays were uncharacteristically sloppy with the ball in the November meeting, making 10 assists while turning the ball over 15 times. Did the Aztecs’ defense play a role in that? Of course, but they were defending a very different team than what we are seeing now.

With Ty-Shon Alexander playing with confidence and taking care of the ball at an elite rate over the past month-plus, I don’t think it’s a stretch to say we could get average Creighton rates in this rematch. Not the Bluejays’ best performance, but something that more resembles their mean.

For the season, they average roughly 5.3 field goal attempts per turnover and carry a 1.4 assist-to-turnover rate. Given that they took 52 shots in this game, the numbers would project 9.8 turnovers, and with 9.8 turnovers, we’d assume 13.7 assists given their season rate. In this game, they averaged 2.2 points per assisted bucket, so now that we are adding 3.7 assists to their ledger, we are bumping up their point total by eight points (notice that I am NOT subtracting any points from the SDSU side despite the Aztecs forcing fewer turnovers, so the difference could even be higher if one fewer leak-out bucket occurs).

Shooting (+21 points)

Despite taking five more 3-pointers than SDSU, Creighton was outscored by 21 points from behind the arc. Twenty-one! Anything can happen in a one-game sample, but that’s not exactly a likely outcome given that the Bluejays average more triples per game than SDSU.

For the season, the Bluejays are shooting 38.5% from 3 and the Aztecs allow opponents to make 28.4% of their attempts from downtown. Average out those rates and expecting Creighton to knock down 33.5% of its 3-pointers makes sense. Repeat the same process for SDSU and the expected shooting percentage is 34.1%. Rounding in favor of SDSU in both cases, understanding the shot attempts could have been slightly impacted by the game flow, both teams would expect to make seven 3s, meaning Creighton makes three more than it did in the first matchup and SDSU makes four fewer.

Was any of that math unreasonable? Forget unreasonable; if you are projecting a game, would these median expectations not be the basis for your opinion of how it would play out? Well, with little effort, I just swung the difference in this game by 37 points. In a completely logical manner, I made three tweaks that could very easily happen and now Creighton wins by six (interestingly enough, my exact projection at the moment should these teams play) instead of losing by 31.

The point of that exercise is not to say that the Bluejays should have won that game in November. They shouldn’t have. They were terrible. Or that they’d be a lock to win if the stars align for an April meeting. Who honestly knows? But knowing what we know now, how could you reasonably assume that the result three months ago would be predictive of a future matchup?

And this isn’t me preaching from a Creighton soapbox, the Bluejays just happen to be the example from the look-ahead bracket. It’s a case of updating opinions and adjusting to what we are currently seeing versus what we saw months ago. That’s not a knock against San Diego State in this instance; it’s buying the ceiling of one team (Creighton here) that we are in the midst of seeing over the current form of another.

I’m asking you to avoid using a one-game sample to make a sweeping claim, especially when that game came in November. I’m now going to leave you with some long-range context, as I am here to provide bracket insight, not to defend my projections from last week.

Over the past two seasons, Final Four teams have won 86.8% of their games played through December and 88.8% of their final 10 games pre-March Madness. That’s right. Despite the rigors of conference play, over the past two seasons, the successful tournament teams have been winning at a higher rate at the end of a long, grueling season against conference opponents than they did early on in a portion of the schedule that features fresh legs and some cupcake matchups. That wasn’t the case over the decade prior (87.7% win rate through December, 76% in their final 10 games) and I think the recent trend is worth buying into as it seems teams have shifted their focus to peaking down the stretch more now than ever.

Source: Read Full Article