Icechicken's 2017 Law School Rankings Forum

[Rigo]

I really don't get what the point is in posting a ranking without posting the criteria/methodology? Right now it looks like "best job placement in the UP" or something.

[icechicken]

The message I'm getting is that I overrated Michigan. But I think the ordinal nature of the rankings exaggerates the extent to which I did that. Here are the raw numbers:

Yale - 27.0
Michigan - 11.1
Harvard - 7.0
Penn - 7.0
Berkeley - 5.1
Cornell - 5.0
Stanford - 2.3
Chicago - 2.0
Duke - 0.5
NYU - 0.1
NU - 0.1
UVA - 0.0
Columbia 0.0

Hopefully things are clearer now.

[Nebby]

Why is a 0L ranking law schools?

[cavalier1138]

The message I'm getting is that I overrated Michigan. But I think the ordinal nature of the rankings exaggerates the extent to which I did that. Here are the raw numbers:

Yale - 27.0
Michigan - 11.1
Harvard - 7.0
Penn - 7.0
Berkeley - 5.1
Cornell - 5.0
Stanford - 2.3
Chicago - 2.0
Duke - 0.5
NYU - 0.1
NU - 0.1
UVA - 0.0
Columbia 0.0

Hopefully things are clearer now.

Yes, the arbitrary numbers you put after schools definitely cleared things up.

But as an NYU student, thank you for putting Columbia where they belong. Nice to see a ranking system that finally puts them in their place.

[A. Nony Mouse]

The message I'm getting is that I overrated Michigan. But I think the ordinal nature of the rankings exaggerates the extent to which I did that. Here are the raw numbers:

Yale - 27.0
Michigan - 11.1
Harvard - 7.0
Penn - 7.0
Berkeley - 5.1
Cornell - 5.0
Stanford - 2.3
Chicago - 2.0
Duke - 0.5
NYU - 0.1
NU - 0.1
UVA - 0.0
Columbia 0.0

Hopefully things are clearer now.

How on earth does a different set of completely unexplained numbers clear things up? Without metrics this is completely pointless. Especially given that based on this list your metrics look really weird. I’d guess you were weighting cost of living in the school’s location brutally heavily but that doesn’t really explain Virginia.

ETA: I get that you’re probably trying to avoid having your methodology savaged by the peanut gallery here but the alternative is to be mocked for not providing a methodology.

[TLSModBot]

Here is some more substantive comparisons to go with icechicken's ordinal numbers to help put them in perspective (don't worry, not giving away your precise intricacies!):

Yale - Three smiley faces
Michigan - one smiley face and a duck
Harvard - Turkey (animal)
Penn - the sound of falling rain
Berkeley - *shrug*
Cornell - broken bone
Stanford - Turkey (country)
Chicago - white privilege
Duke - gangrene
NYU - non-alcoholic beer
NU - Two thumbs down
UVA - Three thumbs down and a frowny face
Columbia - literally murder

ETA: outlandish rankings are presumably based on idiotic methodologies until proven otherwise. At the very least offer.some kind of clarification on what or how you're separating "objective results" from reputational effects and how that practically matters for applicants or their likely outcomes

[Nebby]

I rank my law schools based on most megaposters

1. CLS

[TLSModBot]

Funny I get a similar result when I rank by most sweatshirt worn by Nebby

[Nebby]

1. CLS
2. NU
3. GULC
4. NYU
5. HLS
6. SLS

[cavalier1138]

The message I'm getting is that I overrated Michigan. But I think the ordinal nature of the rankings exaggerates the extent to which I did that. Here are the raw numbers:

Yale - 27.0
Michigan - 11.1
Harvard - 7.0
Penn - 7.0
Berkeley - 5.1
Cornell - 5.0
Stanford - 2.3
Chicago - 2.0
Duke - 0.5
NYU - 0.1
NU - 0.1
UVA - 0.0
Columbia 0.0

Hopefully things are clearer now.

How on earth does a different set of completely unexplained numbers clear things up? Without metrics this is completely pointless. Especially given that based on this list your metrics look really weird. I’d guess you were weighting cost of living in the school’s location brutally heavily but that doesn’t really explain Virginia.

ETA: I get that you’re probably trying to avoid having your methodology savaged by the peanut gallery here but the alternative is to be mocked for not providing a methodology.

I'm strongly reminded of a grade I got back in undergrad for a 100% subjectively graded class. The teacher literally gave a string of numbers at the top of the page, with some separated in parentheses, and then put "=B+" at the end of the line. No explanation, no math, just a line of numbers (that didn't add up to anything between 85 and 90) and a letter grade at the end of the non-equation.

At the very least, I think we've determined that the scale goes from 0 to 30.

[BrainsyK]

I will confirm if anyone is able to reverse-engineer the system.

I'll take a shot. Can you confirm if this was your system?

https://www.amazon.com/Deluxe-Wire-Cage ... B000F0GXTM

[cavalier1138]

I will confirm if anyone is able to reverse-engineer the system.

I'd like to reverse-engineer. Can you confirm if this was your system?

https://www.amazon.com/Deluxe-Wire-Cage ... B000F0GXTM

Dammit. I clicked on that, and now Amazon thinks I'm in a retirement home.

[sev]

Yale - 27.0
Michigan - 11.1
Harvard - 7.0
Penn - 7.0
Berkeley - 5.1
Cornell - 5.0
Stanford - 2.3
Chicago - 2.0
Duke - 0.5
NYU - 0.1
NU - 0.1
UVA - 0.0
Columbia 0.0.

With the exception of UPenn, Cornell, Duke, and UVA the numbers seem to correlate pretty well with the population of the city where the school is located.

Yale - 129k
UMich - 120k
Harvard - 110k
Penn - 1.56m (outlier)
Berkeley - 121k
Cornell - 30k
Stanford - 1m (San Jose)
Chicago - 2.7m
Duke - 263k
NYU - 8.5m
NU - 2.7m
UVA - 46k
Columbia - 8.5m

[Rigo]

Stanford is its own census designated place, tyvm.
Lumping it into San Jose when it doesn't even border San Jose is very, very rude.

[sev]

Stanford is its own census designated place, tyvm.
Lumping it into San Jose when it doesn't even border San Jose is very, very rude.

So I noticed, but I've never visited and figured it was separate in name only. In any case, I've always associated Stanford with San Jose, so there we are.

[icechicken]

Yale - 129k
UMich - 120k
Harvard - 110k
Penn - 1.56m (outlier)
Berkeley - 121k
Cornell - 30k
Stanford - 1m (San Jose)
Chicago - 2.7m
Duke - 263k
NYU - 8.5m
NU - 2.7m
UVA - 46k
Columbia - 8.5m

Amazing. This warrants further analysis IMO

[KENYADIGG1T]

Dean Z doin the most outchea on these streets

[181plz]

So Cornell is 50x better than NYU and infinite x better than Columbia. This makes my admissions decision much easier than I anticipated. Thanks for the help!

[TheBlueDevil]

So Cornell is 50x better than NYU and infinite x better than Columbia. This makes my admissions decision much easier than I anticipated. Thanks for the help!

Apparently Michigan is 111 times better than NYU. So I guess if you're willing to pay sticker for NYU, you should also be willing to pay $370 million for Michigan.

ETA: anyone know if Ms. DeVos will loan me $370 million?

[KENYADIGG1T]

So Cornell is 50x better than NYU and infinite x better than Columbia. This makes my admissions decision much easier than I anticipated. Thanks for the help!

Apparently Michigan is 111 times better than NYU. So I guess if you're willing to pay sticker for NYU, you should also be willing to pay $370 million for Michigan.

I have a feeling only Dr. Evil can pay Yale sticker

[icechicken]

In consideration of the feedback I have received, I had adjusted my formula slightly. Here is a provisional new list of best universities in the T13, with the total score and breakdown for each school included in the format (Score: FC, TC, NL, FC2). That first value, FC, is carried over from the old formula and I feel that it really forms the backbone of my list's effectiveness, although Michigan and, to a lesser extent, Northwestern and Stanford have probably punched a little too far above their weight by focusing on that subscore at the expense of other institutional priorities.

Sadly, this adds an input (the all-important "NL" factor) to the calculation, which before had been purely outcome-focused, but this metric can still be objectively measured, so I do not feel I have compromised this project too much by doing so. On the whole this system is a good deal more intricate but only somewhat less precise.

1) Yale (4260: 27, 77, 25, 2)
2) Columbia (4223: 0, 37, 83, 15)
3) Stanford (2559: 2, 126, 49, 0)
4) Harvard (2505: 7, 51, 49, 1)
5) Chicago (2450: 2, 0, 90, 0)
6) NYU (2385: 0, 37, 24, 12)
7) Cornell (2377: 5, 78, 47, 0)
8) Penn (2190: 7, 53, 26, 3)
9) Berkeley (2161: 5, 54, 47, 0)
10) Michigan (1618: 11, 2, 20, 0)
11) Duke (419: 0, 16, 11, 0)
12) UVA (418: 0, 27, 7, 0)
13) Northwestern (396: 0, 0, 11, 1)

Good news for Columbia!

[sparkytrainer]

In consideration of the feedback I have received, I had adjusted my formula slightly. Here is a provisional new list of best universities in the T13, with the total score and breakdown for each school included in the format (Score: FC, TC, NL, FC2). That first value, FC, is carried over from the old formula and I feel that it really forms the backbone of my list's effectiveness, although Michigan and, to a lesser extent, Northwestern and Stanford have probably punched a little too far above their weight by focusing on that subscore at the expense of other institutional priorities.

Sadly, this adds an input (the all-important "NL" factor) to the calculation, which before had been purely outcome-focused, but this metric can still be objectively measured, so I do not feel I have compromised this project too much by doing so. On the whole this system is a good deal more intricate but only somewhat less precise.

1) Yale (4260: 27, 77, 25, 2)
2) Columbia (4223: 0, 37, 83, 15)
3) Stanford (2559: 2, 126, 49, 0)
4) Harvard (2505: 7, 51, 49, 1)
5) Chicago (2450: 2, 0, 90, 0)
6) NYU (2385: 0, 37, 24, 12)
7) Cornell (2377: 5, 78, 47, 0)
Penn (2190: 7, 53, 26, 3)
9) Berkeley (2161: 5, 54, 47, 0)
10) Michigan (1618: 11, 2, 20, 0)
11) Duke (419: 0, 16, 11, 0)
12) UVA (418: 0, 27, 7, 0)
13) Northwestern (396: 0, 0, 11, 1)

Good news for Columbia!

Stop. This whole thing isn't even funny anymore.

[A. Nony Mouse]

Yeah, at this the point the thread is basically a troll. You need to justify it or I’m locking it.

[BlendedUnicorn]

Yeah, at this the point the thread is basically a troll. You need to justify it or I’m locking it.

It set you up for the best joke I've ever seen you make on TLS so that's something.

[sev]

My attempt at cracking the methodology works for the very top and bottom bands (except stanford I suppose).

1. Harvard 1636/1817
2. Yale 1701/1824
3. Columbia 1754/1858
4. Penn 1740/1850
5. Michigan 1817/1859
6. Duke 1838/1868
7. Berkeley 1868/1894
8. Cornell 1865/1887
9. Chicago 1890/1902
10. Northwestern 1851/1859
11. NYU 1831/1835
12. Virginia 1819/1819
12T. Stanford 1893/1893

(Rankings based on the number of years between the founding of the university and the founding of the law school--which are the two numbers beside each school.)