NESCAC 2014

Started by Becks, February 27, 2014, 08:56:22 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Becks

Well, daughter #3 is following in the footsteps of daughter #1, so I'm looking forward to what will hopefully be another 4 years of NESCAC women's soccer. Lots of time before the first game for me to catch up on what's been happening in the league since daughter #1's last season in 2010 and to provide all sorts of (probably meaningless) pre-season analysis.

Becks

#1
College soccer team roster sizes typically range from the low 20s to the mid 30s. Here's where the NESCAC's stand.

Rank. School - 4-year average (2013, 2012, 2011, 2010):

1. Amherst - 28 (28, 28, 28, 28)
2. Colby - 26.5 (27, 26, 25, 27)
3. Conn - 25.75 (28, 27, 24, 24)
4. Bowdoin - 25.25 (25, 25, 25, 26)
4. Williams - 25.25 (27, 27, 23, 24)
6. Bates - 25 (24, 23, 27, 26)
6. Middlebury - 25 (28, 25, 24, 23)
8. Hamilton - 24.25 (25, 21, 25, 27)
9. Tufts - 24 (23, 24, 24, 25)
9. Wes - 24 (26, 22, 23, 25)
11. Trinity - 22.25 (25, 22, 24, 18)

Observations:
1. There does not appear to be much correlation between team success over the years and roster size.
2. Could be just chance due to the small sample size, but the only team that seems to have a fixed roster size is Amherst.
3. Although the roster sizes of the other teams bounce around a bit, they generally stay within a limited range.
4. No team has had a roster of more than 28 players.
5. There was only 2 occurrences of a roster of fewer than 22 players.

Possible factors affecting roster sizes:
1. Coach preference. Some coaches may like to have lots of players on their roster (i) to make sure they have enough for 11v11 drills/scrimmages even if there are lot of injuries or other attrition and/or (ii) to increase competition for playing time. Other coaches may like smaller roster sizes so they don't have to choose between (i) having a lot of players on the bench that are unhappy with their playing time and (ii) giving playing time to a lot of weaker players.
2. Talent pool. A team that has a deeper pool of talent show up at tryouts is probably more likely to keep a larger roster because they will feel confident that they can play everyone on the roster without a large drop-off in talent.
3. AD issues. On the one hand, there is a marginal cost to the school for each additional player; on the other hand, more girls on the team helps balance out the football numbers a bit for Title IX purposes.
4. NESCAC rules? I'm not sure, but there may be a NESCAC rule that limit the number of players on the team (or the number that travels) to 28.
5. Bus size? Teams may not want to take more girls on the roster than the buses can take. Perhaps the bus limit is 28?

Consequences of roster size. As every player or parent knows, the biggest issue with larger roster sizes is playing time. The larger the roster, inevitably, the more players who are going to get minimal playing time.


Becks

#2
For comparison, here are the 2013 roster sizes at a sampling of other D1 and D3 schools.

D1:
North Carolina - 36
Colgate - 29
Stanford - 28
Notre Dame - 28
UNH - 27
Elon - 27
Wake Forest - 27
Bucknell - 26
Dartmouth - 25
Yale - 25
Ohio State - 25
Holy Cross - 23
Florida State - 22

D3:
MIT - 32
Brandeis - 30
Union - 30
Claremont - Mudd - Scripps - 30
William Smith - 29
Gettysburg - 28
Swarthmore - 27
Pomona-Pitzer - 26

Based on this very unscientific sampling, with a few notable exceptions (e.g. UNC - 36), most D1 schools seem to have roster sizes that are in the range of NESCAC roster sizes, although perhaps averaging slightly larger than the NESCAC average. On the other hand, a lot of top non-NESCAC D3 schools have larger roster sizes than any of the NESCACs.

Becks

#3
Roster size is largely a function of (i) the size of the incoming class and (ii) attrition rate. Here are the size of the incoming freshman classes for the last 7 years.

Rank.School - 7-yr ave (roster for fall of '13, '12, '11, '10, '9, '8, '7) (NA means the roster info was not available.)

1. Conn - 11.0 (9, 12, NA, NA, NA, NA, NA)
2. Trinity - 9.7 (9, 11, 9, NA, NA, NA, NA)
3. Bates - 8.3 (8, 6, 6, 11, 8, 12, 7)
4. Bowdoin - 8.1 (9, 9, 8, 7, 10, 7, 7)
5. Amherst - 7.1 (7, 6, 8, 10, 6, 7, 6)
5. Colby - 7.1 (7, 5, 8, 9, 5, 8, 8)
5. Wes - 7.1 (7, 7, 8, 10, 7, 6, 5)
5. Williams - 7.1 (8, 8, 9, 4, 7, 4, 10)
9. Hamilton - 7.0 (7, NA, NA, NA, NA, NA, NA)
10. Tufts - 6.4 (8, 4, 6, 10, 4, 8, 5)
11. Middlebury - 6.1 (11, 3, 7, 5, 9, 2, 6)

Observations:
1. Some coaches (Bowdoin, Amherst, Colby) seem to like to have pretty much the same size frosh class every year. Something in the 5-9 or 6-10 range. Other coaches (e.g. PK at Middlebury) seem to be happy to let the size of the class fluctuate wildly (2-11).
2. The sweet spot for most schools seems to be a class size that averages 7-8.
3. Larger class sizes can be offset by smaller prior or subsequent class sizes, more attrition, and fewer lateral additions; smaller classes sizes can be offset by larger prior or subsequent class sizes, lower attrition, and more lateral additions.

Becks

#4
The other major factor determining roster size is attrition rate or, its flipside, retention rate.

Here are the retention averages for the NESCAC teams for the last 4 senior classes. Retention was calculated based on the number of frosh on the roster that were still on the roster as seniors. So for example, if there was a frosh class of 8 and 4 of those were on the roster as seniors, the retention rate was 50%. I couldn’t calculate retention rates for Conn, Hamilton or Trinity because their old rosters don’t seem to be available on their websites.

Rank. School – retention average (class of ’14, class of ’13, class of ’12, class of ’11)

1. Middlebury – 95% (5/5, 9/9, 1/2, 5/5)
2. Williams – 84% (2/4, 7/7, 3/4, 9/10)
3. Amherst – 76% (6/10, 6/6, 6/7, 4/6)
4. Tufts – 67% (6/10, 3/4, 6/8, 3/5)
5. Colby – 63% (7/9, 3/5, 4/8, 5/8)
6. Wes – 54% (3/6, 4/7, 4/10, 4/5)
7. Bowdoin – 42% (3/7, 3/10, 3/7, 4/7)
8. Bates – 37% (5/11, 3/8, 5/12, 1/7)

Retention rates are affected by coaches cutting players and players quitting.

Cutting players:
1. Coaches' roster management philosophy. Some coaches may intentionally take larger recruiting classes, give everyone a long look, and then weed out the players they conclude will not be able to contribute. Others may like to take smaller recruiting classes, with the expectation that they will keep almost all of them on the team until graduation.
2. Depth of recruiting class talent. A coach who believes that his/her recruiting class is solid throughout, with little drop off in talent, may be less inclined to cut players than a coach who believes there is a noticeable drop off in talent within his/her recruiting classes. This could explain why the more successful teams have higher retention rates -- more depth in recruiting class and fewer marginal players to cut.
3. Coaching changes. A new coach generally has less loyalty to existing players and vice versa. Bowdoin and Bates are the only schools list above that have had coaching changes in the past 4 years and also have the lowest retention rate. Then again, maybe the coach leaving reflected a general lack of success and/or unhappiness on the team, so actually both the coaching change and the lower retention rates were the result of the same problems.

Players quitting:
1. Playing time. Lack of playing time is the #1 reason that players quit a team. Any player who is not either a starter or a regular sub may question whether what they are getting out of being on the team is worth what they are putting into it. Teams who have rosters with fewer players who are only occasional subs (either because the team has a smaller roster or because the coach regularly uses more subs in games) are likely to have higher retention rates.
2. Player assurance of more playing time with seniority. Players who are not happy with their playing time well decide not to come back the following year unless they believe their playing time will improve substantially the following year. As a result, players on teams who have assurance (based on what the coach or players say or what they see) that their playing time will increase in future years will probably be less likely to quit due to lack of playing time than players on teams who have no assurance of increased playing time in future years.
3. Other factors affecting player happiness on team. While playing time is perhaps the biggest factor affecting player happiness, any other factor affecting player happiness with their experience on the team (eg team success and team chemistry) will also affect the likelihood that players will want to stay on the team.

Becks

#5
Even though the time periods for the recruiting class figures and retention rates in the two posts above are not identical, there is a fair amount of consistency between them. For example, Middlebury, which has the highest retention rate also has the smallest average recruiting class size, while Bowdoin and Bates, which have the lowest retention rates also have among the largest recruiting class sizes.

Very hard to say whether retention rates drive recruiting class size or recruiting class size drives retention rates. High retention rates can cause a coach to take smaller recruiting classes, while low retention rates can cause a coach to take larger recruiting classes. On the other hand, large recruiting classes can cause/allow a coach to cut players (to free up room for the next recruiting class) and lower retention rates, while smaller recruiting classes can cause/allow a coach to keep everyone and raise retention rates.

Becks

#6
Here are the retention rates so far for the classes of '15 and '16. No Hamilton because only the 2013 roster is available.

Rank. School – retention average (class of ’16, class of ’15)
1. Amherst - 100% (5/5, 7/7)
2. Wes - 93% (6/7, 8/8)
3. Williams - 89% (7/8, 9/10)
4. Conn - 83% (10/12, NA)
5. Colby - 77% (3/5, 7/8)
6. Bowdoin - 76% (7/9, 6/8)
7. Bates - 75% (5/6, 9/12)
7. Trinity - 75% (10/11, 15/20)
9. Middlebury - 70% (2/3, 5/7)
10. Tufts - 60% (3/4, 3/6)

A smaller and less complete set of data than the prior retention figures, but still interesting. The retention percentages are higher than the prior data set, but that would be expected since this data set does not follow the players all the way through their senior year. Perhaps most significantly, there is not a lot of consistency with the prior rankings. This may show that retention rates for schools vary from class to class based on the characteristics of that class, circumstances of the team, or just plain random factors, rather than holding consistent for particular coaches or schools. Or maybe it just shows the problems of small sample size.

2xfaux

Amazing.  Good luck to daughter #3!

amh63

FYI.....Amherst's has just brought in a new assistant...div 1 player from same college as the Head Coach.  Info can be found on Amherst website.
  The cycle continues as the Class of '18 will be finalized very soon. HS Spring break is on and the main quad has groups of potential applicants on tour.  Amherst  applicant  pool has again exceeded well over 8,000 and continues to lead the conference in the lowest admit percentage.  Target class size is in the low 400.

See from the tables provided in the Beck posts that Amherst's squad numbers are average for the conference, while enrollment is in the lower have.  Amherst seeks to be competitive in all sports.

Becks


Becks

A partial and tentative 2014 game schedule is available (kind of) at http://www.nescac.com/sports/wsoc/2014-15/schedule . The schedule is clearly not complete (eg the Wes v Tufts and Wes v Trinity games are not on the schedule) and has not been officially "published" on the NESCAC website yet, but I suspect that most of what is there is correct.

Becks

#11
Since I have an HS son as well, I took a look at NESCAC men's and women's soccer stats to see what the similarities and differences were.

Many of the key stats are surprisingly similar.

Median team goals per game average (conf games only) - 2013, 2012, 2011
M - 1.1, 1.2, 0.9
W - 1.3, 1.2, 1.3
The women actually scored slightly more per game, but, given the small sample size, the difference may not be significant.

Median team shots per game average (conf games only) - 2013, 2012, 2011
M - 12.6, 12.3, 11.8
W - 12.4, 11.4, 13.5
Very similar for men's and women's teams.

Saves percentage of #6 (approx median) ranked keeper (conf games only) - 2013, 2012, 2011
M - 0.800, 0.814, 0.791
W - 0.802, 0.821, 0.836
Fairly similar, but surprisingly (to me) the median women's keeper had a slightly higher saves percentage.


Some stats, however, do show differences.

Median corner kicks per game (conf only) - 2013, 2012, 2011
M - 4.4, 5.0, 4.4
W - 3.8, 3.9, 3.7
I think this difference is real. One possible theory is that women are not coached to seek corner kicks as much because women are generally weaker headers of the ball, so corner kicks are less valuable in the women's game than the men's game.


The biggest statistical difference that I found, however, was in fouls.

Median team fouls per game (conf only) - 2012, 2011, 2010 (I included 2010 because I could not find 2013 stats)
M - 12.2, 12.2, 10.5
W - 7.9, 8.5, 8.6
So a typical NESCAC women's game has only about 2/3 the number of fouls called as a typical NESCAC men's game, which, subjectively, seems about right to me.

Median team yellow cards per season (conf only) - 2013, 2012, 2011
M - 13, 8, 9
W - 0, 1, 2
This is a huge difference. There were an average of about 2 yellow cards handed out in each men's game, while a women's team might see 2 all year.  Under FIFA rules, this would mean that NESCAC men commit about 10x as many "reckless", "unsporting" or "deliberate" fouls as NESCAC women.  I suspect men do commit more of these fouls, but I think part of it is also a general reluctance on the part of refs to hand out cards in women's games even when the conduct deserves it.


Becks

#12
D3 recruiting class information is always spotty, but here's what I've found so far for the incoming recruiting classes:

College - Name - Home Town - Club Team - Position

Amherst - Delancey King   - Braintree, MA   - Scorpions SC ECNL U18   - MF
Bates - Sarah McCarthy - ? - East Meadow SC Dynamite NPL U18 - GK
Bowdoin - Rachel Stout - Lakeville, MN - Shattuck-St Mary’s Sabres U18 - GK
Tufts   - Chandler Quintin - MA - Scorpions SC ECNL U18 - MF
Williams - Natasha Albaneze - Northport, NY - East Meadow SC ECNL U18 - F
Williams - Hanna Kaeser - Pelham, NY - Connecticut FC ECNL U18 - MF
Williams - Katherine Sands - Rye, NY - Connecticut FC ECNL U18 - DM
Williams - Jacqueline Simeone   - Milton, MA - IMG Soccer Academy - D

The information is obviously very incomplete, but I do think the fact that I located 4 Williams recruits suggests that Williams's recruiting class may be strong relative to other NESCACs. The argument is that it is probably somewhat likelier that D3 recruits on ECNL teams or other high level teams will have lots of D1 teammates and will not want to feel left out when their D1 teammates have their commitments posted.

I know Wes has an incoming class of 8-10 (depending on whether recruits who were accepted regular decision decide to come and whether wait-listed recruits get in), but I don't want to post any names until they otherwise become public, probably when the preseason roster comes out in August.

Becks

#13
Looks like there has been a general league-wide increase in roster sizes over the past 8 years.

School (alphabetically) - roster size in '13, '12, '11, '10, '9, '8, '7, '6
Amherst - 28, 28, 28, 28, 23, 25, 21, 22
Bates - 24, 23, 27, 26, 26, 24, 24, 23
Bowdoin - 25, 25, 25, 26, 26, 25, 23, 23
Colby - 27, 26, 25, 27, 24, 25, 24, 24
Conn - 28, 27, 24, 24, 23, 22, 24, 22
Hamilton - 25, 21, 25, 27, 25, 26, 26, 27
Middlebury - 28, 25, 24, 23, 25, 24, 26, 27
Trinity - 25, 22, 24, 18, 19, 19, 19, 19, 17
Tufts - 23, 24, 24, 25, 19, 23, 20, 19
Wesleyan - 26, 22, 23, 25, 22, 22, 25, 24
Williams - 27, 27, 23, 24, 24, 23, 25, 22

Median roster size by year:
2013 - 26
2012 - 25
2011 - 24
2010 - 25
2009 - 24
2008 - 24
2007 - 24
2006 - 23

In 2006, only 2 of 11 NESCAC teams had more than 24 girls on their roster. By 2013, 9 of 11 NESCAC teams had more than 24 girls on their roster.

I think the most likely reason for the increase in roster sizes is that the number of solid female soccer players seeking to be recruited and play in college has been steadily increasing. Back in the mid-2000's, at some NESCAC schools, the end of the roster contained a good number of players who had not played club soccer and may not have actively sought to be recruited for soccer. I think those types of players are harder to find on NESCAC rosters now. With a larger number of solid players available, coaches feel more comfortable keeping more players on the roster. Coaches probably also have a harder time cutting players who have actively sought to be recruited, have communicated with the coach for an extended time, and are obviously very committed to playing soccer in college.

Becks

#14
Updated list below. New addition in bold.

Amherst - Delancey King   - Braintree, MA   - Scorpions SC ECNL U18   - MF
Bates - Sarah McCarthy - ? - East Meadow SC Dynamite NPL U18 - GK
Bowdoin - Rachel Stout - Lakeville, MN - Shattuck-St Mary’s Sabres U18 - GK
Trinity - Maddie Snyder - Montclair, NJ - Match Fit ECNL U18 - M
Tufts   - Chandler Quintin - MA - Scorpions SC ECNL U18 - MF
Williams - Natasha Albaneze - Northport, NY - East Meadow SC ECNL U18 - F
Williams - Hanna Kaeser - Pelham, NY - Connecticut FC ECNL U18 - MF
Williams - Katherine Sands - Rye, NY - Connecticut FC ECNL U18 - DM
Williams - Jacqueline Simeone   - Milton, MA - IMG Soccer Academy - D