It seemed like a useful exercise to apply the proposed competitiveness index to the map I created with purerly geographic rules.
Note that my rules permit counties to be split more than two ways in order to minimize counties that are split, and minimize the total number of fractional counties. This is not permitted under Ohio law and happens once in Stark county. This would be easy to fix, and would have no substantial change on the results.
I've completed Ohio using the usual rules.
Only one city, Columbus, is larger than the population of a district. Columbus also has a number of unincorporated pockets inside, and I'm only guessing that I can make a district entirely within the City limits, and provide the necessary links between the pockets and the adjacent districts. With that assumption, rules P-1,2,3 are all zero.
OH has a number of counties with large urban centers, and that forces some county splits. I split Montgomery and Stark that were less than one district size, and split Stark into three pieces to avoid
splitting a third small county. Rule P-4 is two.
All three large counties had a district wholly within them, and Cuyahoga has two wholly within. Rule P-5 is zero. All three also only have two partial districts. Rule P-6 is 11 partial districts, and rule P-7 is three maximum in any county.
I only used data from the 2004 Persidential election, rather than the six individual races called for by the proposal. Nonetheless it gives some idea. I've also tries to number my districts in a similar fashion to the current CDs.
The proposal uses the 3 closest races over the previous 8 years (4 elections). The Ohio Secretary of State has the results from the previous 3 elections: 2 presidential elections, 2 senatorial elections, and the state executive elections in 2002. The 2 presidential elections have been closest; the senate elections weren't close at all; the gubernatorial election was not close, so the 3rd closest was a down ticket race won by about 57-43. Republicans could benefit by losing a minor executive race because it would make the districts appear to be more competitive.
The current CD-plan, which has a 12:6 Republican majority, has a competitive index of -8. It has 4 competitive districts, but they were all carried by Bush, including one that is represented by a Democrat (Strickland). It has 8 uncompetitive districts, but only one balanced pair. Surprisingly for a GOP-favoring plan, this is because the 4 Republican seats are more one-sided than 3 of the Democrat seats (as in your plan, the Cleveland district is overwhelmingly Democrat).
By making the uncompetive GOP seats slightly more balanced, and the uncompetitive Democrat seats even more so I can balance 3 pairs. I then can either force a 5th Democrat seat into the uncompetive zone, or bring one of the 4 GOP seats to just under 15%.
I can tip a couple of the the competitive seats to the Democrats favor, giving me 2 balanced competitive pairs, and make a currently 5.5% GOP seat "competitive". This then gives a competitive index of +7, without really changing the competitive balance. Two of the GOP representatives (Chabot and LaTourette) were elected in 2000 when their districts at the time were carried by Gore.
There may be a ladder effect in any case, where the uncompetitive districts for one party are more uncompetitive than those of the other party. For example under the current CD plan, the districts are:
D 63.4%
R 30.8%-D 25.6% 0.2% (shift to balance)
R 29.3%-D 16.6% 7.7%
R 27.7%-D 16.6% 6.7%
R 22.4%-D 10.9% 6.5% (including 4.1% Democrat)
R 14.7% (no change)
R 14.7% (no change)
It would be interesting whether a computer could find a plan with an index of 31 (8 balanced competitive pairs, one not-balanced competitive district and Cleveland).
Lets say find If I could get a Cuyahoga suburban district that was 4.99% Democrat. It ought to be possible to get your NE district within 5%. Split Summit county, and run a pair of districts towards the southwest. Rearrange the Canton and Youngstown districts, and place the Steubenville-Wheeling in a district stretching to the west. The Sandusky and Toledo districts can be paired. Make the Franklin and Hamilton county seats 4.99% Republican, which frees up Democrats for balancing with the more rural areas. I'm ending up with too many Republicans in the west.
So maybe try for one intermediate 14.9% Republican district, reducing the score to 30. Or 7 balanced competitive pairs, one other competitive district, a pair of balanced uncompetitive districts, and Cleveland, for a score of 27. That would be pretty hard to beat with any reasonable plan.