MBB: NESCAC

[amh63]

For those interested......Amherst is still scheduled to play at Rhode Island College tonight starting at 7:30 PM.  You can obtain video/live stats of the game via the Amherst website.....and probably the RIC website...RIC has a fine web broadcast that was free in past years.
Info is not yet available on D3hoops.com.   Did notice that a number of schedule games were postponed...due to weather conditions?

[nescac1]

Seems like the Amherst game was finally postponed -- I don't see how they possibly could have played.

LOTS of weather postponements this year (Midd-Hamilton still needs to be played, for example).  But really, it is getting late in the year ... at this point, might some of these games just not be played at all?  So long as they are non-conference games, is losing one game from the schedule really going to affect anything?

Although I imagine Amherst might push hard for this game, considering that the Jeffs have very few road games to begin with this year (I think seven including RIC) which might hurt their seeding come tourney time, unless of course they finish undefeated in the regular season. 

[Hugenerd]

Seems like the Amherst game was finally postponed -- I don't see how they possibly could have played.

LOTS of weather postponements this year (Midd-Hamilton still needs to be played, for example).  But really, it is getting late in the year ... at this point, might some of these games just not be played at all?  So long as they are non-conference games, is losing one game from the schedule really going to affect anything?

Although I imagine Amherst might push hard for this game, considering that the Jeffs have very few road games to begin with this year (I think seven including RIC) which might hurt their seeding come tourney time, unless of course they finish undefeated in the regular season.  

Dont forget! Going undefeated at home is 2.33 times worse than going undefeated on the road! So if they had the exact same schedule as Midd or Williams, but had all home vs. all road, their SOS would be 3/7 of the all-road schedule. What a system!!!

[nescac1]

Wow, that is ridiculous, I didn't know the disparity was so great.  That seems to give WAY too much weight to road games.  But given that, yeah, Amherst definitely wouldn't want to lose one of its few road games, especially vs. an above .500 opponent like RIC. 

[amh63]

Thanks nescac1......went out to lunch with the missie.....and my evening plans are gone!
Now I understand why the Amherst WBB team coach was determined to get two away games  when his team missed out on the Vegas games.......not that it really matters based on the WBB play to date.
Understand that the first regional rankings come out today.....by the NCAA?  Should be interesting and wonder if they are up to date, etc.

[Pat Coleman]

Regional rankings come out on Wednesdays, late in the afternoon. We'll have them here.

[magicman]

Wow, that is ridiculous, I didn't know the disparity was so great.  That seems to give WAY too much weight to road games.  But given that, yeah, Amherst definitely wouldn't want to lose one of its few road games, especially vs. an above .500 opponent like RIC.  

nescac 1
Over on the Pool C board, KnightSlappy has posted his blog site where he has been doing regional rankings. Yesterday he posted his latest set and for the Northeast Region he ranks 14 teams which is 3 more than the NCAA ranks. Amherst wasn't listed in his latest rankings and I thought that was odd so I sent him a message asking if their strength of schedule and RPI were so bad that they didn't make the list despite a 17-0 record. This is what he wrote back to me.

"magicman,
Amherst's wSOS is coming out as 0.400 which is 11th worst in D3. I'm guessing the committee will rank them high because of the 17-0 record (and rightfully so), but the RPI calc doesn't respect the really low SOS.

As of now, their wRPI puts them at #15 in the region, one off of my list.

It's really their home/road mix (very home heavy) that's screwing them, but they'll see a late season SOS boost when they finish off at Williams and at Middlebury."

In view of these facts I would think they would want to make up that game with Rhode Island, because as you stated it's not only a road game, it's one against a decent team.

Here is a link to KnightSlappy's rankings.  Just scroll down a bit and you'll find the latest rankings.

http://tomaroonandgold.blogspot.com/search/label/Regional%20Rankings

[amh63]

Glanced over the "prediction" of Knightslappy's regional rankings and basically think it is very strange and "broken"
First, it seems that as the season advances and Amherst wins more (therefore winning percentage remains 1.00), they fall lower and lower and out of the rankings.  What is strange, is the appearance of teams.....conference teams  that Amherst has beaten.....rising upward.  Bowdoin, Trinity, and even Conn......appearing.  Of course there is also Elms.  How can such teams rise up as they lose.....I guess it is SOS (?) and other factors such as color of uniforms, inverse ratio of margins of points scored or some other calculation.  Does the calculators ever take a step backwards and look at strange trends to see if they made a bad calculation.....like a wrong placement of a decimal point?  Oh well, I see number crunchers that are blind and cannot see mistakes appearing even if they looked for them.  

[lefrakenstein]

Wow, that is ridiculous, I didn't know the disparity was so great.  That seems to give WAY too much weight to road games.  But given that, yeah, Amherst definitely wouldn't want to lose one of its few road games, especially vs. an above .500 opponent like RIC.  

nescac 1
Over on the Pool C board, KnightSlappy has posted his blog site where he has been doing regional rankings. Yesterday he posted his latest set and for the Northeast Region he ranks 14 teams which is 3 more than the NCAA ranks. Amherst wasn't listed in his latest rankings and I thought that was odd so I sent him a message asking if their strength of schedule and RPI were so bad that they didn't make the list despite a 17-0 record. This is what he wrote back to me.

"magicman,
Amherst's wSOS is coming out as 0.400 which is 11th worst in D3. I'm guessing the committee will rank them high because of the 17-0 record (and rightfully so), but the RPI calc doesn't respect the really low SOS.

As of now, their wRPI puts them at #15 in the region, one off of my list.

It's really their home/road mix (very home heavy) that's screwing them, but they'll see a late season SOS boost when they finish off at Williams and at Middlebury."

In view of these facts I would think they would want to make up that game with Rhode Island, because as you stated it's not only a road game, it's one against a decent team.

Here is a link to KnightSlappy's rankings.  Just scroll down a bit and you'll find the latest rankings.

http://tomaroonandgold.blogspot.com/search/label/Regional%20Rankings

Does that mean Amherst is unlikely to receive a Pool C bid, even if they finish with only 1 or 2 losses? Seems like that would be ridiculous.

[Pat Coleman]

I would say that seems unlikely but they may be seeded lower than other teams with more losses if the trend holds true.

[lefrakenstein]

The Massey SOS numbers currently has Amherst's schedule played ranked as the 172 toughest in the country, which is indeed awful.

But Massey also gives the predicted SOS for the full schedule, and Amherst should finish a respectable 73rd. Williams, in comparison will finish 59th and Middlebury 149th.

Absent of the home/road split, Amherst I think actually has the hardest schedule of the three. They play traditional regional contenders like RIC and Brandeis and Babson. Hamilton is probably the hardest non-con on Williams schedule.

Amherst seems to be getting screwed b/c this was their year to host the PVC (1 year in 4 they host), plus they have 5 home games in the NESCAC this year, plus the home/away with Brandeis was at home this year, and now the RIC game has been cancelled.

Where does KnightSlappy have the Panthers ranked in region? Their SOS is little better than ours...

I would go to the site myself, but I'm in China and the gov't blocks all blogspot sites.

[amh63]

Lefrakenstein.......I decided to provide you some numbers from KnightSlappy's charts.....and to support my earlier posts.....
The chart dated 1/24/11 covering games through Jan. 23.. had Williams first and Midd. 2nd. ....Trinity was ranked 5th and Bowdoin was ranked 10th.....followed by Amherst at 13th.
The latest chart dated 1/31/11 had Williams first again and Midd 2nd.....Bowdoin rose to 6th and Trinity fell to 9th with Conn appearing at 14th ahead of Amherst which someone pointed out to 15th.
Trinity must have fell due to a lost to Amherst?
This past weekend Amherst easily beat Bowdoin.  I do not care if the win was at Amherst.  In a game, one can tell that one team is much better than another....whether it is on a neutral court, away or at home.  Amherst (I am bias) could play Bowdoin 5 times and I would state without hesitation that Amherst would beat Bowdoin every time.
To have Conn. ranked ahead of Amherst or even appearing is also very strange.  Conn, Trinity and even a team like Elms which is ranked 4th...are teams that I feel Amherst can beat anywhere (Conn was beaten by Amherst in Ct. with a depleted team).

To think that Bowdoin who may end near the bottom third of the conference.....would have a better chance of getting a Pool C bid over Amherst is not correct......but one cannot challenge the "numbers"

[Hugenerd]

You have to remember a few things with regards to Knightslappy's rankings:

1) They are not the official NCAA regional rankings.

2) He is using an RPI calculation to rank the teams.  That means he is only using a single, purely mathematical criteria.  RPI = 1/3 * (winning percentage) + 2/3 * (SOS).  He is plugging in the NCAAs formula for SOS.  However, the NCAA does not use a pure RPI, they do not tell us how they weight WP, SOS, and the other primary criteria.  Therefore, the official NCAA rankings could be very different than the RPI numbers.

3) The new weighted SOS formula is horrible.  To demonstrate, lets play a little game.  Which of the following 4 games helped Amherst's RPI the most (calculated using total D3 records so far, I didnt take the time to account for in-region games only): Anna Maria, Williams, Conn College, or Wesleyan?

And the answer would be: Anna Maria! In fact, Williams had the least SOS value out of the 4 games listed.  Why you ask? Because they played Williams at home.  Thats right, beating a 9-11 Wesleyan team on the road (SOS = (9/19)*1.4 = .66) counts more for SOS than beating an otherwise undefeated Williams team at home (1*.6 = 0.6).  Conn College also a bigger game with regards to SOS (0.5*1.4= 0.7).  All of which are dwarfed by Anna Maria (.667*1.4=.933). 

So, although the Williams win was nice, the 3 biggest games for Amherst this year so far, with respect to the new weighted SOS calculation, have been Anna Maria, Conn College, and Wesleyan.

Therefore, if you want a high SOS, all you need to do is play mediocre teams from bad conferences who will have ~0.500 record.  As long as you play those games on the road, you get the extra 1.4 multiplier, which gives you a decent ~0.7 SOS.  Obviously you cant play all your games on the road, but if you could, why play anyone difficult?  Playing a 0.400 team on the road counts exactly the same as playing a 0.933 team at home.

So if team A and B both played a 3 game season, and Team A won all 3 of their games at home against Middlebury, Williams, and West Conn (combined 93% winning percentage), while Team B won all 3 games on the road at Mount Ida, Newbury, and Norwich (combined 43.8% WP), Team B would actually have a higher RPI because both teams have 100% WP and Team B has higher SOS (.438*1.4=0.613 >> 0.93*.6 = 0.558).  What a system! I say we bring back the QOWI!

[amh63]

Thanks Hugenerd.....for the info/effort.   People who make these calculations and equations and weighing factors for BB games must be following teams/games a little......or are they just "blind" to BB and grind out numbers with little notice of possible errors to input data or new weighting factors.  Elms college is an example of your post.  Elms lost to both Williams and Amherst early in the season.  Amherst beat Elms at home by 33 and Williams won by double digits.  Elms has at least 6 losses and is in a relative "Weak" conference this year.  Yet Elms is ranked 4th in the 1/31/11 regional rankings by KnightSlappy.  It would put Elms in a better position for a Pool C bid then WPI and/or West Conn.  You would think the rise of Elms would bring some notice to the blogger to question his calculations.
Anyway...thanks again.   Hope your MIT degree is near.  Best of Luck with that.   What "course" are you in at MIT?

[Hugenerd]

Thanks Hugenerd.....for the info/effort.   People who make these calculations and equations and weighing factors for BB games must be following teams/games a little......or are they just "blind" to BB and grind out numbers with little notice of possible errors to input data or new weighting factors.  Elms college is an example of your post.  Elms lost to both Williams and Amherst early in the season.  Amherst beat Elms at home by 33 and Williams won by double digits.  Elms has at least 6 losses and is in a relative "Weak" conference this year.  Yet Elms is ranked 4th in the 1/31/11 regional rankings by KnightSlappy.  It would put Elms in a better position for a Pool C bid then WPI and/or West Conn.  You would think the rise of Elms would bring some notice to the blogger to question his calculations.
Anyway...thanks again.   Hope your MIT degree is near.  Best of Luck with that.   What "course" are you in at MIT?

Like I said before, I think when you add in other factors, like the other "primary criteria" the NCAA uses, some common sense will prevail.  In other words, I dont think Elms has a better shot at a Pool C than WPI, West Conn, or Amherst.  Their position on the RPI list is simply due to skewing caused by the weighted SOS.  The NCAA does not calculate an RPI, no one, to my knowledge, has ever published a report telling us how they weight each of the primary criteria. In the end, there is a panel of people doing these rankings, and the NCAA selections, so if they feel the weighted SOS is not a good indicator, they could put more weight in other factors, like overall winning percentage, record versus ranked opponents, or head to head.  We wont know for sure, but I wouldnt lose any sleep over Elms, personally.

Hope your MIT degree is near.  Best of Luck with that.   What "course" are you in at MIT?

Thanks.

Course X, ChemE, with a focus in biomedical and drug delivery applications, defending this Spring.