Koerner’s 63-Game Mega Bracket Breakdown: How to Gain a Big Edge on Your Pool
- Sean Koerner uses his power ratings to break down every game in the bracket -- all 63 of 'em!
- By comparing his proprietary metrics to the actual picks being made by users, it's easy to spot specific team that offer increased value.
Before I dive into my bracket, I wanted to point out that I will be filling in my selections with the help of our Bracket Simulator, which is powered by my power ratings.
Within the simulator, you see the optimal bracket combinations for your pool size, get projected spreads for every possible matchup and see where the futures betting market is off according to my simulations.
For the purposes of this article, I’ll be using the Upset Level of 3 as the default, since that will mirror the actual expected volatility of the tournament.
It can be very useful to play around with the various Upset Levels in the tool and hit “Simulate” just to remind yourself that nothing is certain. There will likely be some significant upsets early on and some of the top seeds may get knocked out early. However, you’ll also notice that we usually see the better teams prevail and make it to the Final 4.
Before filling out your bracket, you should check out Jonathan Bales’ excellent piece on overall strategy for winning your bracket pool, our instant breakdowns of each first round matchups and a cheat sheet for each region.
My goal with this article is to walk you through setting my initial bracket using the strategy of making picks by weighing the expected win probability of each matchup but also what percentage of users appear to be picking each team to advance (according to Yahoo! Pick distribution).
This will help us highlight spots to strategically take an underdog that is more likely to win than most people are giving them credit for. We are not trying to be “perfect,” but instead put ourselves in situations that are “smart contrarian.” That allow us to mostly take teams that are heavily favored, but pick our spots to go against the grain when there’s an edge in doing so.
For all potential matchups, I will be using the Matchup Tool in the Simulator, which spits out the expected spread/win probability for any potential matchup in the tourney via my very own power ratings. These numbers are a great way to help set your bracket by getting to see exactly just how close some of the potential matchups could be.
Data from Yahoo as of Monday evening; you can find updated numbers here.
#1 Duke vs. #16 NC Central/North Dakota State
- Win %: Duke 99%
- Users: Duke 99%
No need to waste any brain power here. Just take Duke.
#8 VCU vs. #9 UCF
- Win %: UCF 50.4%
- Users: UCF 55%
The betting market has this as at around a pick‘em, and we need to treat it as such. Big picture, this pick has pretty low leverage as we will just pick Duke to handle whichever team wins here. I’m going to simplify my decision and say: If VCU’s Marcus Evans suits up, I’ll go with the Rams; if he is ruled out, I’m going with UCF.
#5 Mississippi State vs. #12 Liberty
- Win %: Miss St. 77%
- Users: Miss St. 73%
A lot of bracket veterans are going to make it a point to pick one or two 12-seeds to pull off an upset. I would think that is why the market is a bit bullish on Liberty here. I’m locking in Mississippi State here, as there isn’t enough of an edge to go smart contrarian quite yet.
#4 Virginia Tech vs. #13 Saint Louis
- Win %: VT 86%
- Users: VT 89%
There is a bit of an edge in taking Saint Louis here: There are a higher percentage of brackets having them advance here than the Billikens’ odds of winning. But I’m simply going to roll with the chalk here with Virginia Tech and take my chances on a more likely upset with more leverage over the field.
#6 Maryland vs. #11 Belmont/Temple
- Win %: Maryland 66% (vs. Belmont), Maryland 77% (vs. Temple)
- Users: Maryland 81%
We have our first leverage play of the East Region, as users aren’t wanting any part of the winner of the play-in game to see who will be the #11 seed. Ideally, Belmont would be the Terrapins opponent. And if that’s the case, I would strongly consider taking the underdog in that spot, especially if 75%+ of brackets are sticking with Maryland.
#3 LSU vs. #14 Yale
- Win %: LSU 82%
- Users: LSU 88%
I’m taking LSU here and using a bit of a look-ahead strategy. Since I’m considering an 11-seed to upset, I’ll want to mitigate that risk a bit by locking in LSU to advance. You can see in the Round of 32 section below how my plan for the next round will unfold based on the result of the 11-seed play-in game.
#7 Louisville vs. #10 Minnesota
- Win %: Louisville 64%
- Users: Louisville 68%
My power ratings had Louisville only favored by 4 in this matchup. And despite the line opening at -5.5 and Louisville receiving 63% of the action so far, the line has dropped to -5 in some spots.
I’ve already locked in Minnesota +5 and will likely have them pull off the upset here and take them in my bracket. There’s a bit of leverage here with users taking Louisville in nearly 70% of brackets and we have a bit of a safety net with the winner having to face Michigan State next.
Based on 10,000 simulations we can expect a median of 45 for the total sum of all the East seeds that advance to the Round of 32. By taking Minnesota to upset here, it would bring my East Region sum of seeds in the next round to 44. I think this shows I have an ideal blend of smart contrarian plays to advance to the next round.
#2 Michigan State vs. #15 Bradley
- Win %: Michigan St. 98%
- Users: Michigan St. 98%
Take the Spartans.
Round of 32
#1 Duke vs. #8 VCU
- Win %: Duke 96%
- Users: Duke 99%
With nearly 50% of users taking Duke to win the title, we have to seriously consider fading them to go all the way. We need to be strategic as to when we have them lose (if we so choose), but this is way too early and does not meet the criteria of being “smart contrarian.”
#5 Mississippi State vs. #4 Virginia Tech
- Win %: VT 53%
- Users: VT 72%
A huge advantage of our March Madness Bracket Simulator is the Matchup Tool that allows us to see what the expected spread/win probability would be for all potential matchups. Users are giving Virginia Tech way too much credit here and we need to capitalize on it.
Should Virginia Tech be favored here? Yes. Do the Hokies have a slightly easier path to get to this matchup? Yes. However, at a nearly 70% pick rate in this spot, we have a 15-20% edge over the field. Take Mississippi State here as you only need a little luck to gain a few points on the field.
#6 Maryland or #11 Belmont vs. #3 LSU
- Win %: LSU 50.4% (vs. Maryland), LSU 71% (vs. Belmont)
- Users: LSU 69% (vs. Maryland), LSU 93% (vs. Belmont)
This part of the bracket offers the most potential scenarios, considering I’ll be waiting for the result of the 11-seed play-in game to determine my strategy. If Belmont wins, I will strongly consider taking it to upset Maryland in the first round and then advance LSU to the Round of 16. If Temple wins the play-in game, I will lock in Maryland to advance to the second round.
After seeing nearly 70% on LSU to beat Maryland in what would essentially be a pick ‘em, I’m tempted to just have Maryland advance to the Round of 16 no matter what the outcome of the play-in game may be.
I think it’s critical to never constrain yourself to earlier round picks when you spot a potential leverage play, especially when the bracket scoring is doubled. That 20% edge is too tempting to pass up here so I’m leaning toward Maryland here and hoping that the Terps face Temple to make the path to get here that much easier.
#10 Minnesota vs. #2 Michigan State
- Win %: Michigan State 86%
- Users: Michigan State 97%
Part of taking #10 Minnesota to upset #7 Louisville is that I’m conceding this matchup to Michigan State. The idea is to build up to the possible scenario of Michigan State being the team to knock Duke out in the Elite 8. Having scenarios that in a sense makes the Spartans’ path easier to get there are part of my plan.
If you took Louisville to play them here it’s worth pointing out that we’d give the Cardinals a 25% chance of upsetting the Spartans compared to only 11% of users taking them in that spot. It’s an interesting spot to gain a bit of leverage over the field with a strategic bracket-busting pick, but I’m not going there.
Round of 16
#1 Duke vs. #5 Mississippi State
- Win %: Duke 91%
- Users: Duke 99%
One thing we need to keep in mind when we are this deep into the bracket is to remember the path each team needed to get here. Mississippi State has roughly a 36% chance of making it this far, so this is not the time to get cute and have them upsetting Duke. Remember, we would have already cashed in on our leverage taking them to upset Virginia Tech in the previous round. It’s time to take our chips off the table.
#6 Maryland vs. #2 Michigan State
- Win %: Michigan St. 77%
- Users: Michigan St. 92%
Keeping in mind the “path to get here,” I’m going to stick with the chalk here and go Michigan State. More importantly I am doing so with the possibility of making them the team that knocks Duke out.
Round of 8
#1 Duke vs. #2 Michigan State
- Win %: Duke 66%
- Users: Duke 85%
While I would say most of the country has this matchup in their bracket, what I’m going to do is something that only 13.5% of people seem to be willing to do: I’m taking Michigan State to upset Duke here and advance to the Final 4.
Looking at our Final 4 probabilities in the Bracket Simulator, we see that the Spartans make it about 24% of the time, giving us a 10%+ edge over the field.
#1 Virginia vs. #16 Gardner-Webb
- Win %: Virginia 99%
- Users: Virginia 99%
Don’t even think about it. Take Virginia.
#8 Ole Miss vs. #9 Oklahoma
- Win %: Ole Miss 50.4%
- Users: Oklahoma 53%
If 60% of users were on Ole Miss, I would likely go with Oklahoma here. Considering slightly more people are on the 9-seed, I’m going to roll with Ole Miss here.
#5 Wisconsin vs. #12 Oregon
- Win %: Wisconsin 62%
- Users: Wisconsin 58%
This is the most intriguing 5/12 matchup to me. I was actually expecting users to be fairly split on this, but with nearly 60% on Wisconsin I’m thinking that a lot more savvy users will be on Oregon, which is white hot and has covered the spread in eight straight games. With that in mind, I’m thinking there is a bit more leverage in taking the slight favorite here in Wisconsin.
#4 Kansas State vs. #13 UC Irvine
- Win %: Kansas St. 71%
- Users: Kansas St. 83%
In higher stakes pools or pools with a higher percentage of sharp brackets, you’ll probably see UC Irvine being selected closer to the 35% implied win probability in the betting markets.
Kansas State has a key player, Dean Wade, who is expected to miss. Factoring this in, I have Kansas State as 6-point favorites, but the betting market is even more bullish on UC Irvine, with the line at +4.5 now. Any way you slice it, we have a leverage opportunity here by taking UC Irvine.
It’s also worth noting that in all the other regions, I am getting an average of ~45 for the sum of all seeds that advance to the next round, but the South has an average of 47, indicating that upsets are more likely to come in this region in the first round.
#6 Villanova vs. #11 Saint Mary’s
- Win %: Villanova 66%
- Users: Villanova 83%
The high percentage of brackets will have Villanova, which makes Saint Mary’s tempting. I think whichever team advances here, I will have losing to Purdue, which makes me more inclined to go with the 11 seed and try to possibly steal a point from the field.
#3 Purdue vs. #14 Old Dominion
- Win %: Purdue 92%
- Users: Purdue 91%
This is an odd scenario where our expected win percentage is slightly higher than the user pick percentage. I’m thinking that could mean Purdue will offer some possible value to upset the 2-seed Tennessee further along in the bracket. We will see if that’s the case a bit later.
#7 Cincinnati vs. #10 Iowa
- Win %: Cincinnati 53%
- Users: Cincinnati 64%
We have another great leverage spot here with users being a bit too heavy on Cincinnati. Give me Iowa. I actually had Cincinnati only being a -1.5 favorite in this spot, so I’m also going to be on Iowa +3.5 here as well.
#2 Tennessee vs. #15 Colgate
- Win %: Tennessee 97%
- Users: Tennessee 98%
Take Tennessee. This would give me a sum of all seeds advancing at 53. That is well above the average of about 46 you should shoot for in each region. If you took the averages of my East/South regions, it would be exactly 46. We are operating on a pretty optimal amount of favorites vs. upsets at this point.
Round of 32
#1 Virginia vs. #8 Ole Miss
- Win %: Virginia 86%
- Users: Virginia 97%
Take Virginia. One thing I’m very cognisant of is the heavy amount of upsets I have in the first round of this region. This would, in theory, make the path for the top teams that much easier. I’m going to possibly be a bit more conservative the rest of the way in this bracket, but let’s see if user pick percentage forces me to stay aggressive…
#5 Wisconsin vs. #13 UC Irvine
- Win %: Wisconsin 82%
- Users: Wisconsin 85%
I love having this matchup for a couple reasons. 1) your average player will have the chalky #4 Kansas State vs. #5 Wisconsin 2) your advanced player is likely to go #12 Oregon vs. #13 UC Irvine here.
I wanted to have more of a hybrid approach to give myself a bit of an edge. Of the other combos, 78% have Oregon beating UC Irvine, while 59% have Kansas State beating Wisconsin.
Needless to say, I have quite a bit riding on Wisconsin to win here. Only 29.5% of brackets have them making it to the Round of 16, while I’m estimating them to have closer to a 39.9% chance.
#3 Purdue vs. #11 Saint Mary’s
- Win %: Purdue 83%
- Users: Purdue 89%
What appears to be a bit of a leverage opportunity needs to be taken with a grain of salt. We have to remember that 83% of users are taking Villanova to advance to this matchup. I would say that users who had Saint Mary’s advancing are a bit sharper, hence they realize it’s time to get off the ride and take Purdue here.
Of the users who went with the #6 Villanova vs. #3 Purdue matchup, 51% actually took Villanova to win that matchup. Using our Matchup Tool in the Bracket Simulator, you can see that my power ratings would have Purdue being -4.5 with a 66% chance of winning.
We are being handed an edge in this part of the bracket to take a #3 seed team to advance, as expected, so we need to take advantage of it.
#2 Tennessee vs. #10 Iowa
- Win %: Tennessee 84%
- Users: Tennessee 95%
Round of 16
#1 Virginia vs. #5 Wisconsin
- Win %: Virginia 79%
- Users: Virginia 93%
I already have enough riding on Wisconsin to get here, so I’m not going to be tempted by the the +14% edge between our expected win probability and the percentage of brackets with Virginia advancing here. Give me the Cavaliers, and we’ll see if there is a potential edge in having them get upset in the next round.
#2 Tennessee vs. #3 Purdue
- Win %: Tennessee 53%
- Users: Tennessee 78%
There appears to be a massive edge in taking Purdue here. If we set up this game in the Matchup Tool, you can see that Tennessee would only be a 1.5-point favorite here. The Vols most certainly have the easier path to get here, but if the matchup does end up being the #2 vs. #3 we have to take the +25% of edge in Purdue.
Taking a step back, I have Purdue being around 33% to make it to the Elite 8 vs. just 16% of brackets having them making it there. We do have to take on a bit of risk in making these less-than-likely picks, but the payoff is much greater than it should be. This is what helps us shoot up the standings in our bracket pool.
Round of 8
#1 Virginia vs. #3 Purdue
- Win %: Virginia 66%
- Users: Virginia 87%
While I’m tempted to stay on the Boilermaker Special all the way to the Final 4, I think now is the time to get off the train. Virginia advances to the Final 4 in 50% of our simulations, but only 43% of the brackets currently have them making it that far.
Users appear to be overweight on Tennessee with 31.6% of brackets in the Final 4 when we have them closer to a 23% of making it. Remember that we aren’t trying to make the “perfect” bracket, but instead put ourselves in spots to capitalize on market inefficiencies.
Taking Purdue over Tennessee, then Virginia over Purdue is a very safe way to do just that.
#1 North Carolina vs. #16 Iona
- Win %: UNC 99%
- Users: UNC 99%
#8 Utah State vs. #9 Washington
- Win %: Utah St. 50.4%
- Users: Utah St. 54%
I had the opening spread for this matchup pegged at Utah State -1. They opened up at -3, and despite receiving 69% of the action so far, the line has stayed there. I’m taking Washington +3 and making them my pick here since I think it’s much closer to a true pick’em.
#5 Auburn vs. #12 New Mexico State
- Win %: Auburn 77%
- Users: Auburn 82%
New Mexico State could be a pretty popular upset pick the more users look into their 30-4 record and current 19-game win streak. Unfortunately it was against schools like Grand Canyon, Chicago State and Texas A&M International. I’m not seeing this as a spot to take on the risk, so I’m going to play it safe and go with Auburn.
#4 Kansas vs. #13 Northeastern
- Win %: Kansas 81%
- Users: Kansas 93%
When I have a team expected to win over 80% of the time, I really need a +15% edge or so to really consider taking the upset. I’m going to roll with Kansas here.
#6 Iowa State vs. #11 Ohio State
- Win %: Iowa St. 66%
- Users: Iowa St. 72%
I peaked ahead and noticed that there appears to be some value in having Iowa State advance quite a bit in this part of the bracket. Let’s have them hanging on to win a close game against a tough #11 seed in Ohio State and pick them to advance here.
#3 Houston vs. #14 Georgia State
- Win %: Houston 88%
- Users: Houston 93%
Sticking with my extremely chalky favorites to advance in the Midwest Region and going with Houston here. I’ve been finding it easier to gain an edge against the field in the next few rounds, and that’s where we can really start to rack up some points against the field.
#7 Wofford vs. #10 Seton Hall
- Win %: Wofford 56%
- Users: Wofford 50%
Wofford has been on an absolute tear of late, going 11-1 against the spread in its last 12 games. They are another team that has benefited from a very easy schedule and went 0-3 against top 25 teams earlier in the season. I think we have to go with Wofford here if 50% of the brackets are going with Seton Hall. Wofford opened as 2.5-point favorites and that has moved up to -3.5 despite there being 60% of the tickets on Seton Hall so far.
#2 Kentucky vs. #15 Abilene Christian
- Win %: Kentucky 99%
- Users: Kentucky 99%
Round of 32
#1 North Carolina vs. #9 Washington
- Win %: UNC 88%
- Users: UNC 98%
UNC is the weakest 1-seed in the field, but I do not think it is worth rolling the dice here. My numbers give Washington a 12% chance to beat UNC, but we need to remember that the Huskies have to win a 50/50 matchup in order to even get here. It’s not worth betting on a 6% event to occur when you aren’t getting nearly enough in return vs. the field.
#5 Auburn vs. #4 Kansas
- Win %: Auburn 50.2%
- Users: Kansas 56%
Kansas is entering the tourney having lost its second-leading scorer, Lagerald Vick, due to a leave of absence. The Jayhawks should be fine in advancing to this round, but I have Auburn as the slight favorites here. Let’s take full advantage of the slight edge given to us here and roll with Auburn to move on to play UNC.
#6 Iowa State vs. #3 Houston
- Win %: Iowa St. 50.4%
- Users: Houston 67%
As I mentioned earlier, having Iowa State advancing to the Round of 16 presents an edge we need to exploit. Houston has the better chance of getting to this matchup, but if these two teams square off, I actually have Iowa State being a 1-point favorite. This is another example of how to leverage our Matchup Tool when filling out your bracket.
#7 Wofford vs. #2 Kentucky
- Win %: Kentucky 98%
- Users: Kentucky 92%
Round of 16
#1 North Carolina vs. #5 Auburn
- Win %: UNC 70%
- Users: UNC 93%
This is a spot I feel offers some leverage in having a 1-seed get beat. UNC would likely be -5.5 favorites here. Eighty-one percent of users have the Tar Heels advancing to the Elite 8, while my sims only make it 61%.
Auburn has a much better chance (18.2%) than the 5.2% of brackets that will take the chance on them here. If you are in a pool of, say, 200-plus entries I would certainly roll the dice with Auburn here. If your pool is fewer than 200 entries, I may take UNC to advance to the next round to then get eliminated.
Either way, the Tar Heels will be the 1-seed I’m a bit aggressive in having an earlier exit than the other three.
#6 Iowa State vs. #2 Kentucky
- Win %: Kentucky 73%
- Users: Kentucky 88%
Considering I have the 1-seed getting into trouble and failing to advance to the Elite 8, I’m not going to get cute here and will just slot Kentucky into my Elite 8 with the intention of them being in my Final 4 and beyond.
#1 North Carolina vs. #2 Kentucky
- Win %: Kentucky 70%
- Users: Kentucky 90%
Only 22% of brackets have Kentucky in the Final 4 despite our Simulator having them get there 39% of the time, so I’ll have the Wildcats advancing here.
If you’re in a smaller pool and put UNC here, one important note: My power ratings would make this game a pick’em, while the market is overrating the Tar Heels.
It’s really odd to see that 59% of brackets have UNC make it to the Final 4 (second-highest rate in the tourney), despite the Tar Heels reaching in our Simulator 37% of the time. That’s a ton of value. The betting market reflects a similar outlook for UNC: It has the worst odds for a 1-seed to make the Final 4, while Kentucky has the best odds for a 2-seed.
Round of 64
#1 Gonzaga vs. #16 Fairleigh Dickinson/Prairie View A&M
- Win %: Gonzaga 99%
- Users: Gonzaga 99%
The play-in game has zero impact here. Take Gonzaga no matter which team it faces here.
#8 Syracuse vs. #9 Baylor
- Win %: Syracuse 50.4%
- Users: Syracuse 66%
The heavy pick percentage of Syracuse is sort of forcing my hand to take Baylor. This is a dead-end pick anyway, as the winner is likely going to be a big underdog against Gonzaga in the next round.
My numbers make Syracuse a 1-point favorite in this game; the market currently has them at -2.5. I’m waiting to see how far that number moves and then I will likely throw some action behind Baylor on the spread, as well.
#5 Marquette vs. #12 Murray State
- Win %: Marquette 66%
- Users: Marquette 61%
I’m going to take Marquette here and end up taking all four No. 5 seeds to advance. Earlier I mentioned how 5/12 is a popular matchup for your average user going with one or two 12-seeds pulling off the upset.
This is actually a sound strategy, according to the Bracket Simulator:
Should I just go back real quick and change one of my 5-seed victories to a 12-seed? Not so fast. Here’s why: Yes it’s true that it’s more likely that exactly one No. 12 seed wins. But this doesn’t necessarily mean that taking exactly one is the optimal strategy per se.
What I am doing this year is taking all 5-seeds to win the first round and three of the four to advance to the Sweet 16. The edge I have been seeing with the market overweighting 4-seeds is what’s driving this.
Getting those teams correct is worth double, so why would I go out of my way to play it a bit safer to rack up 1 point when I can have a bigger edge over 2?
#4 Florida State vs. #13 Vermont
- Win %: Florida St. 86%
- Users: Florida St. 90%
Take Florida State.
#6 Buffalo vs. #11 Arizona State/St. John’s
- Win %: Buffalo 66%
- Users: Buffalo 83%
Whichever team Buffalo faces, I have them being 4.5-point favorites. The market is a bit too high on the 31-3 Bulls and think this is a good spot to take the underdog. I’m guessing most people rather just lock in Buffalo here and not pick a TBD team to have to worry about later.
I think it’s worth taking the dog here considering I will be selecting Texas Tech no matter which team emerges from this group.
#3 Texas Tech vs. #14 Northern Kentucky
- Win %: Texas Tech 92%
- Users: Texas Tech 95%
Take Texas Tech.
#7 Nevada vs. #10 Florida
- Win %: Nevada 53%
- Users: Florida 52%
Let’s not overthink this relative coin flip here. I have Nevada -1.5, while the betting market has the Wolf Pack up to -2 right now. It appears slightly more people are willing to roll with the 10-seed here, so let’s take the slight edge and grab Nevada.
#2 Michigan vs. #15 Montana
- Win %: Michigan 97%
- Users: Michigan 98%
Round of 32
#1 Gonzaga vs. #9 Baylor
Win %: Gonzaga 90%
Users: Gonzaga 97%
#5 Marquette vs. #4 Florida State
- Win %: Florida St. 62%
- Users: Florida St. 70%
I’m actually going to go with the public and take Florida St. here. This is actually the first 4-seed I have advancing to the Round of 16.
#11 Arizona State/St. John’s vs. #3 Texas Tech
- Win %: Texas Tech 86%
- Users: Texas Tech 96%
Like I mentioned earlier, it doesn’t matter who wins the #11 seed play-in game. I will likely take the winner of that game to beat Buffalo in the Round of 64, before falling to Texas Tech here.
#7 Nevada vs. #2 Michigan
- Win %: Michigan 75%
- Users: Michigan 87%
The +13% edge here is not enough to consider taking the upset here. Remember that we need Nevada to win a coin flip in order to even have this matchup. Give me Michigan here.
Round of 16
#1 Gonzaga vs. #4 Florida State
- Win %: Gonzaga 77%
- Users: Gonzaga 79%
Gonzaga appears to be a 1-seed we don’t want to mess with too much early on. Let’s get the Zags into the Elite 8 and we can find out if there is any value in having them make the Final 4.
#3 Texas Tech vs. #2 Michigan
- Win %: Michigan 53%
- Users: Michigan 73%
While I have been very conservative so far in the West, now is the time to pounce on +20% edge in this matchup. People who are filling out their brackets without the help of our Matchup Tool probably don’t realize that this game would be very close.
I’d have Michigan being right around a 1.5-point favorite. Texas Tech advances here 36.1% of the time in our Simulator, while only 21.9% of the brackets have the Red Raiders reaching it that far.
Meanwhile, a lofty 58% of brackets have Michigan advancing, but our sims only give them a 45.5% chance of doing so.
#1 Gonzaga vs. #3 Texas Tech
- Win %: Gonzaga 66%
- Users: Gonzaga 85%
Hitting this exact matchup would give us enough of an edge to climb up the standings. So while it’s tempting to continue to fade the public consensus, I’m going to select Gonzaga here. We have the Zags reaching the Final 4 in 49.4% of our simulations, and only 44.7% of brackets currently have them doing so.
There’s always a sense of feeling like you are taking too many favorites, or too many underdogs. I’m really viewing most of my picks as “I think they have a better chance of getting to and winning this matchup than the market.”
Before I dive into my Final 4, wanted to share a cool value I pulled from the simulator. The median for total sum of seeds in the final 4 is 8. You can see the actual breakdown here:
As you can see, six is actually the most likely, and that’s what I have my Final 4 at. This is a reminder that it’s OK if your Final 4 looks a bit chalky; the cream does usually rise to the top.
The fact that I don’t have Duke and/or UNC in my Final 4 is where I am being contrarian.
#2 Michigan State vs. #1 Gonzaga
- Win %: Gonzaga 56%
- Users: Gonzaga 63%
We are so deep into the bracket here it’s a bit easy to get locked in on trying to evaluate this exact matchup. I am instead going to think of things in context of my overall strategy.
Michigan State being here is betting against Duke making the Final 4. Because of this I’m going to take Gonzaga to make the Finals. This will give me a little bit of an out: If Duke does in fact make it this far, I can still win if Gonzaga advances and the rest of my bracket does well.
Only 13.5% of brackets have Gonzaga advancing to the Finals, and we have them reaching roughly 24% of the time in our Simulator. We are taking advantage of the edge provided here by selecting the second-most likely team to make the Finals out of this side of the bracket.
#1 Virginia vs. #2 Kentucky
- Win %: Virginia 53%
- Users: Virginia 56%
Both teams offer an edge vs. the field, as our Simulator has Virginia with a 26.6% chance of advancing (compared to 18.2% of brackets) with Kentucky at 22.2% (compared to 14.5% of brackets).
If I filled out 10 different brackets, I would probably have six with Virginia advancing here, and four with Kentucky advancing. For this specific bracket, I’ll go with the more likely outcome and take Virginia.
#1 Gonzaga vs. #1 Virginia
We have both around 14% chance of cutting down the nets. Only 8.7% of brackets currently have Gonzaga and 5.3% have Virginia, so I’m going to grab the most value here by taking UVA to win it all.
It’s always worth remembering that it’s called “March Madness” for a reason. It rarely goes as planned, so we need to take advantage of that. Because Duke is such a clear favorite, there are going to be way too many people taking the Blue Devils to win it all, which creates a ton of value on other teams.
Having said that, I also didn’t spend the last month or so creating as accurate power ratings as I could just to completely throw them out the window and pick underdogs for the sake of picking underdogs!
The median total points for the Finals comes out to 141 points. You can use that along with the Matchup Tool to come up with a pretty good projected score for the final tiebreaker.