2022 Quality Curve Analysis - February Edition

With our second of three looks at the 2022 QC, there's a lot to uncover, so here's what you can expect:

1. How the QC has changed since the JAN QC Analysis
2. How the QC compares to its historical counterparts
3. The bigger picture on quality in 2022
4. My overall thoughts at this point in the season

 

What's Changed since the JAN QC

The chart shows the change pretty clearly: More Steepness. What does this portend?

1. Steepness of the QC implies sanity in the tournament. Since we are comparing 2022 Jan to 2022 Feb, it simply means the tournament (if held today) would be more chalkier than it would have been in Jan. We would need to compare/contrast the Feb curve to other years to get an idea of how many upsets we could be getting from this particular tourney (we will do later).
2. The increase in steepness resulted from a pivot of the QC curve along the 15-spot and the 16-spot. Generally speaking, most teams from the 1- to 15-spots increased their quality while most teams from the 16- to 50-spots decreased their quality.
• In past years, steepness usually happens by teams at the front of the curve (1-25) improving their quality at a faster rate than teams at the back of the curve (26-50). 2021, 2019, 2018 and 2017 all demonstrated this feature (with their own unique exceptions and their own specific changes from Feb to Mar). 2016 is the only QC I've seen that has done it by the QC rotating along a specific point. In the Jan Analysis Part 2, I talked a lot about Inflection Points because I had a hunch that 2022 might do this, but unlike 2016 I wasn't gutsy enough to make the call for it in 2022.
  
• In that same article, I pointed out the inflection points on the full Kenpom Curve, and those spots were 2, 20, 45, 97, and 150. On this chart, the 2-spot has the sharpest drop from its preceding spot than any spot on the QC and it is the only spot from the 1- to 15-spots that is below its Jan value. The 45-spot inflection point is the exact same location on this curve where the Feb QC crosses over the Jan QC and demonstrates higher quality. Finally, the 20-spot happens to be relatively close to the current inflection point at 15/16. All in all, the inflection points discovered in Jan forewarned us as to what Feb could bring.
  
3. Another noticeable feature of this QC is the bowing (a bow-and-arrow shaping, minus the arrow) of the QC. I don't think I have ever talked about bowing of the QC, but I'm sure that I have talked about bowing in the SC and what it implies (better-than-expected performance from the seed group). In the QC, the importance of curve-bowing is different because teams are not seeded according to efficiency ratings (and as a student of advanced metrics, I'm glad they aren't seeded this way). If I had to give any significance of this bowing-effect, I would say it creates the hypothetical tier-list that I described in the Jan QC Part 2. Here's the hypothetical tier-list:
• Tier 1: Team #1 (Above-average 1-seed)
  
• Tier 2: Teams #2-#4 (Average 1-seeds)
  
• Tier 3: Teams #5-#11 (2-seeds)
  
• Tier 4: Teams #12-#15 (4-seeds)
• Tier 5: Teams #15-#21 (5-seeds)
  
• Tier 6: Teams #22-#50 (7- thru 12-seeds)
• Essentially, teams in a tier-group that do not get seeded in their expected group would be under-dogs. Thus, any teams in Tier 3 (six total) that do not get seeded as a 2-seed (only four 2-seeds available) would be formidable challenges to teams on their level. Imagine being a 2-seed facing a 3-seed that has 2-seed quality (this is usually the recipe for 3-seeds that advance past 2-seeds). As always with the QC, it matters where they get seeded and this seeding affects the SC.

A Different Look at the Changes since the JAN QC

This chart uniquely confirms what the previous chart was telling. This is the range of the QC over the duration of the month. I think it tells us two important facts.

1. The majority of teams on the Feb QC (blue line) are playing at either their best or their worst, meaning the QC is in lock step with the maximum (green line) or the minimum curve (red line).
• Best: #1, #4, #9-#11, #13, #46-#50
• Worst: #16-#38 (and this is probably the craziest thing I've seen in a while, more later).
• Mid-point of the range: #2, #5-#7, #39-#42 (Moved to both the minimum and the maximum and currently sitting in the middle of both extremes).
2. The maximum and minimum curves do not form any significant bottle necks (narrow stretches of max/min ranging, meaning the teams haven't made significant directional movements up or down). The only possible area of bottle-necking is around the 33- thru 37-spots. I will keep an eye on this area because the Jan max/min ranges had slight bottle-necking around the 16- thru 19-spots and the 36- thru 38-spots. Ironically, the #16-#19 spots comprise the gap between the current 15/16-spot inflection point and the 20-spot inflection point on the full-length Kenpom curve.

Historical Comparison of the FEB QC

This chart shows how the current QC compares to its historical counterparts (those QCs at their Feb point in time). Here's the breakdown:

• 1- thru 6-spots: Compares to 2018/2021 (Not good for 1- and 2-seeds)
• 7- thru 13-spots: Compares to 2019/2021 (Likely means one to two 2- thru 4-seeds in the F4)
• 16- thru 42-spots: Worse than all QCs (Likely means few double-digit seeds in the R32 and 1- and 2-seeds unlikely to be upset by the S16 opponents)
• 43- thru 50-spots: Possibly some 12-seed magic if seeded correctly and draws a weak 4-/5-seed)

I left out the 14- and 15-spots because they are almost in lock-step with the 2021 QC, but there's not really anything significant to draw from it since our current QC as well as the 2019 QC abruptly deteriorates in that location. Keep in mind that each of the historical QCs underwent significant changes from Feb to Mar, and I think the same will apply to the 2022 QC.

FEB QC: The Bigger Picture

Here's the full-length Kenpom Curve (discussed in Jan QC Part 2 article) for February:

• From the 46- to the 80-spot, quality has significantly improved. This range of teams should feature the 11- thru 13-seeds in the tourney.
• From the 104- to the 129-spot, quality has significantly declined. Historically, three or four teams from this range will earn anywhere from a 13- to a 15-seed.
• Th 179-spot (the median of the KC) stands at -0.96.
• The negative-value of the median tells us that the entire curve is still left-skewed, which means a higher quantity of lower-quality teams pulling down the average.
• In the Jan QC, this spot stood at -0.94, so there has been slight deterioration in the general quality of the curve. I assume this could be interpreted as separation from the top-quality teams and the middle-quality teams, which would further support our hypothesis of increasing tournament sanity.

Final Thoughts

It's not a strong year (nothing like 2017 or 2019), but it's not a weak year (like 2018 or 2021) either. There's still a month to go though, so it can still go either way. I think the biggest impact this year is the transfer portal, with coaching changes holding its own at #2. Both of these factors affect chemistry. Players that know their system and have added responsibilities now have to learn new teammates that are either talented and didn't work out at the previous school or are experienced and have to learn a new system. 50+ new coaches mean 1/7th of the teams have to learn new offensive and defensive schemes. I honestly believe the rotation along the 15/16 spot could be explained by these two phenomenon. The top teams are hitting their stride because they weren't as affected by one or both of these factors, and the lower teams are simply 2-4 weeks behind in development (meaning they could potentially hit their stride in the coming month and we would see a rise in this part of the QC). It is possibly why the top teams are at their maximum and the lower teams at their minimum.

I have also been following various bracketologists and their projections of the tournament field. I used to have my own watchlists, but I abandoned them due to their workload on me and the lack of predictive value received from the work. Anyways, the more I look at them, the more I look in disgust at the lower-seed teams. As of now, the 8-seeds and lower are nothing but a revolving door. Very few have more than one or two quality wins, and one loss usually sends the team down two seed-lines. I don't think I've ever seen a bubble this wide. It absolutely blows my mind that the NCAA thinks the tournament should go to 96 teams and most years they struggle to even field 68 teams! It's also another reason why I don't see any threats to 1- and 2-seeds from the 8- thru 10-seeds (maybe a 7-seed on an overrated 2-seed). Then again, we are dealing with 18- to 22-year old males: One minute, you barely make it, and the next minute, you're slaying giants (2021 UCLA is a good example).

As always, thanks for reading my work, and I will have one more article before the Mar QC Analysis, with the hopes of a second or even third.