Warning: Contains numbers.

Even though coaches don’t like to talk explicitly about replacing routines because every team is different and every gymnast is a unique gemstone with her own personal sparkle blah, blah, blah, barf, finding the routines to replace scores that are no longer with the team can be a major struggle.

Some teams have more work to do than others. As a way of measuring the value of a gymnast that has graduated or retired, we can compare that gymnast’s RQS to the score we would expect from an average replacement, a 7th-8th gymnast on a apparatus who competes in the event of an injury and keeps the team from falling apart. Certainly, that’s going to vary for every team, and Alabama on vault will be expecting a much higher replacement score than Ohio State on beam.

Still, we can assume that on average most of the top teams will be looking at a replacement score of about 9.800. 9.800 is just OK for a team expecting to make Championships, and nearly all of these teams have 9.800s who were sitting on the bench last season.

So, in analyzing how much replacement work the top 12 teams from last year have to do, we can examine how much the average team score would decrease if we replaced the graduated/retired routines from 2012 (measured by RQS) with 9.800s. That is the tenths above replacement score. For example, if all the Ferguson, Stone, and Cindell routines from last year become 9.800s instead, Oklahoma’s team score goes down an average of 0.700.

A few notes:
-Gymnasts are included if they made the final lineup for the team, be it in Super Six or Semifinals. That’s why Cindell is here even though she likely wouldn’t have competed if not for Nowak’s injury. A healthy Nowak makes up some of that 0.700 right away.
-In cases where a gymnast competed on an event but the RQS was below replacement level (below 9.800), I did not include that routine in the statistics as it should be replaceable.
-In cases where a gymnast did not have an RQS because of injury but did contribute in the postseason, I used the postseason average instead. I also intervened in the case of Kyndal Robarts since her very low RQS was not representative of the fact that Utah will have to find a score to replace hers on floor. It’s cheating, but it creates more realistic numbers. What can I say? I’m a rogue. Without that change, Utah’s number would be 0.390, if you’re interested.
-The numbers slightly underestimate the importance of the seniors for Ohio State, where 9.800s were more rare and valuable last season. The Buckeyes have some of the most replacing to do.

Tenths above Replacement:

1. Oklahoma – 0.700
Megan Ferguson – 0.380 (Bars: 9.910, Beam: 9.930, Floor: 9.940)
Sara Stone – 0.295 (Vault: 9.935, Beam: 9.875, Floor: 9.885)
Candace Cindell – 0.025 (Bars: 9.825)

2. Arkansas – 0.545
Jaime Pisani – 0.440 (Vault: 9.910, Bars: 9.895, Beam: 9.895, Floor: 9.940)
Mariah Howdeshell – 0.105 (Bars: 9.905)

3. UCLA – 0.510
Elyse Hopfner-Hibbs – 0.220 (Vault: 9.815, Beam: 9.885, Floor: 9.920)
Aisha Gerber – 0.145 (Bars: 9.890, Beam: 9.855)
Tauny Frattone – 0.145 (Vault: 9.920, Floor: 9.825)

4. Georgia – 0.458 
Kat Ding – 0.338 (Vault: 9.910, Bars: 9.935, Beam: 9.830, Floor: 9.863)
Gina Nuccio – 0.085 (Bars: 9.885, Floor: 9.800)
Laura Moffatt – 0.035 (Beam: 9.835)

5. Utah – 0.457
Kyndal Robarts – 0.222 (Vault: 9.900, Beam: 9.855, Floor: 9.867)
Stephanie McAllister – 0.180 (Bars: 9.865, Beam: 9.825, Floor: 9.890)
Cortni Beers – 0.055 (Bars: 9.810, Beam: 9.845)

6. Oregon State – 0.440
Leslie Mak – 0.305 (Vault: 9.810, Bars: 9.885, Beam: 9.930, Floor: 9.880)
Olivia Vivian – 0.135 (Bars: 9.895, Beam: 9.840)

7. Stanford – 0.393
Alyssa Brown – 0.233 (Vault: 9.908, Bars: 9.875, Beam: 9.850)
Nicole Pechanec – 0.160 (Vault: 9.810, Bars: 9.900, Floor: 9.850)

8. Alabama – 0.310
Geralen Stack-Eaton – 0.310 (Vault: 9.895, Bars: 9.890, Beam: 9.820, Floor: 9.905)

9. Ohio State – 0.240
Alyssa Marohn – 0.080 (Vault: 9.840, Bars: 9.815, Beam: 9.825)
Nicole Krauter – 0.060 (Vault: 9.835, Beam: 9.825)
Taylor Jones – 0.050 (Vault: 9.800, Bars: 9.815, Floor: 9.835)
Casey Williamson – 0.050 (Bars: 9.845, Floor: 9.805)

10. Nebraska – 0.235
Lora Evenstad – 0.235 (Vault: 9.825, Bars: 9.910, Floor: 9.900)

11. Florida – 0.100
Nicole Ellis – 0.065 (Bars: 9.865)
Amy Ferguson – 0.035 (Floor: 9.835)

12. LSU – 0.055
Ashley Lee – 0.055 (Vault: 9.855)

Yesterday I made the offhand comment that there was no statistical advantage to being in the evening session versus the afternoon session at Nationals. This was based entirely on my own perceptions and a reaction against the general narrative that teams should prefer the evening session because scores increase throughout the day. Well, it’s time to put that assumption to the test.

To find how this actually plays out, I looked at the All-Around, which is the only title decided on the first day of competition and therefore is the most relevant to this argument. I took the average of the top 5 All-Around finishers from each session over the last 10 years to see if the evening AAers actually score higher.

Afternoon AA Average – Top 5 Finishers:
2011 – 39.405
2010 – 39.460
2009 – 39.500
2008 – 39.455
2007 – 39.525
2006 – 39.425
2005 – 39.485
2004 – 39.485
2003 – 39.450
2002 – 39.495
Total Average: 39.469

Evening AA Average – Top 5 Finishers:
2011 – 39.385
2010 – 39.495
2009 – 39.540
2008 – 39.545
2007 – 39.490
2006 – 39.440
2005 – 39.595
2004 – 39.600
2003 – 39.470
2002 – 39.495
Total Average: 39.506

Well, there we have it. There does appear to be a slight statistical advantage for AAers competing in the second session. Now, some of these numbers may be skewed by the home team receiving some home scoring in the evening session, but I don’t think that alone is enough to account for the difference. It also should be noted that the differences were larger in the past and, over the last few years, the advantage has been statistically minimal or, as in 2011, inverted. In fact, the last three AA winners have come from the early session, including Kupets’s massive 39.800 in 2009. The only years where we see a truly notable advantage for the second session were 2004, 2005, and 2008.

Bear with me a bit on this one.

I often think about scoring in terms of “positive routines” and “negative routines.”  Positive routines are those scoring over 9.800, routines that would put the team on a positive path toward breaking 196.  Negative routines are those scoring below 9.800, ones that would put the team below 196 if maintained.  A score of 9.800 exactly is basically breaking even.

All of the very best schools will be able to put up very positive routines in the 5th and 6th positions (solid, consistent 9.9-level performances).  Championships are so often decided by the rest of the lineup, not the stars.  I decided to take a look at how the 1-4 performers were scoring for the top teams, and how many of the routines were positive.  Here’s how it breaks down for the top 8 teams in the country.

Percent of #1-#4 Routines Scoring Over 9.800 (Overall)
1. Utah – 60.9%
2. UCLA – 57.5%
3. Alabama – 56.3%
3. Oklahoma – 56.3%
5. Nebraska – 53.1%
6. Arkansas – 45.8%
7. Florida – 45.0%
8. Georgia – 41.3%

It’s no surprise that Utah occupies the top spot on this list as well as the rankings, having recorded three straight scores over 197.  Teams can’t score that high unless they are getting contributions throughout the lineup.

The issue for Arkansas, and one that may come into play more as the season progresses, is that nearly all their top scores come from Grable and Pisani, and even the numbers that put them at 45% on this list often come from Pisani when she is competing in the 4th position.  They have a lot of other 9.8ish gymnasts in their lineups, but those 9.8s need to become at least 9.85s in April.

I was a bit surprised to see Georgia at the bottom of this group (and the clear bottom as well), but they are often relying on Ding and Worley in anchor positions to push their numbers up.  They’re seeing a lot of 9.775-9.800s from Davis, Earls, and Persinger in those earlier spots.  Interestingly, their worst numbers in this respect come from vault and beam.  Didn’t Jay say those would be their strengths?

Individual apparatus numbers after the jump: