I checked the SSD we used for Council. The racks are B-Racks.

By Ken Kazinski (Kjkazinski) on Wednesday, October 16, 2013 - 10:38 am: Edit

It will be a few days before Robert can get the data from the other tourney's.

Does anyone have any data on the win/loss record of the Maesron?

By Ken Kazinski (Kjkazinski) on Wednesday, October 16, 2013 - 09:26 pm: Edit

At Council we discussed it and were going to reduce the TM's to a single type of missile.

I have to re-visit my notes to see which one.

It looks like everyone agrees the shield should be 30.

OK to remove 2 APR.

Who is the keeper of the official unsanctioned SSD?

By Peter Bakija (Bakija) on Thursday, October 17, 2013 - 08:41 am: Edit

Barry has been doing all the upgrades and tweaks.

By Ken Kazinski (Kjkazinski) on Friday, October 18, 2013 - 05:38 pm: Edit

For Barry:

Yes, the TG's have some very nice advantages, but they have some glaring weaknesses that nobody thought to take advantage of during the tournament.


1) At high energy levels, six or higher, the TGs have a reduced range.


Keep in mind, that at energy five, max range is eight, and damage at that range is eight... Not bad... but an overloaded photon at that range has an average damage slightly higher at that range.


At energy six, max range is five and damage at that range is 10... Only slightly better.


At energy seven, max range is three and damage at that range is 12... Photons at that range do more damage with a slightly lower chance to hit.


At range two, the photon has the same chance to hit but significantly higher damage potential.


A range one, max energy goes up to eight and it becomes a photon torpedo.


There are a couple of ways to play this ship on Turn 1 and 2...


First you can start holding four energy per tube and maybe goto five energy per tube and run down the center of the map. This limits you to a range eight shot and you've used three power per tube, two for holding plus an additional one to upgrade. Total of 12 energy into TG... If your opponent stays out of range eight, your holding cost on turn two just went up to 2.5 per tube.


Or you can corner dive on turn one... putting four or five points of power per tube in on turn one and completing with a single point of power on turn two. If you load five per tube on turn one, you've got six point torps on turn two and a max range of five.


The counter is to keep out of range five on turn two. At that point, on turn three, the Maesron is sucking wind. It needs three points of power per tube just to hold the weapons and it gets slower.


At Co5N, not a single person stayed out of range five on turn two when I corner dived on turn one. In fact in many cases, I was able to get a range three shot, and battery up to seven power on turn two.


Note that after that first firing pass, the Maesron, at least at Co5N almost always, had empty phaser caps, empty batteries, and empty tubes.


Either way, the down the center shot or the corner dive, the turn after firing, the Maesron almost always loads one point of power per tube during the recharge turn. This gives it a lot of speed to run away while re-arming. Even so, very often, I didn't have the power to fully recharge the phaser caps on the run away turn.


The down side, is that for the second firing pass, I've got to put almost all of the arming energy into the tube on the firing turn. Yes, disruptors have to do that too, but disruptors are much more power efficient than Tachyon Guns.


At Co5N, on the second firing turn, I'm going to be slow on the second firing pass. Almost always, at Co5N, I was facing an opponent who charged in for the second firing pass... If they had stayed at range and sniped on that turn, I would have been too slow to catch them, and too short range to return the sniping. And on the following turn after that, I'm going to still be slow because I'm holding expensive short range guns.


In a nutshell, here is the achilles heel of the Maesron... The hold cost for the TG is twice that of photons and at higher energy levels where you do a lot of damage, the range is fairly short. If you can guess what the Maesron is doing and stay out of range when the TG are at higher energy levels and close with it when the TG are at lower energy levels... The Maesron is going to suffer.


Batteries can help out a little, allowing you to charge the TG to slightly lower power levels and longer ranges during EAF and then boost them at the point of firing, if you get to closer range.


There is a counter counter tactic to all this... and that is to not load the TG at all on the turn after firing. Than if your opponent decides to snipe on the turn when you should have the TG up, and you put a lot of power in on that turn. Than the following turn after that, when they think your going to be slow holding TG, you'll actually be fast completing your TG.


In other words, think about playing the Maesron as a ship with a direct fire version of a pseudo torp. Those people who play against it knowing this, will find that the Maesron is not really that powerful.


As for reducing the power to 38 and the front shield to 30... I'll look into that after the Co5N tournament is over... Don't want to switch ships in mid stream.


Best Regards,


Barry Kirk

By Peter Bakija (Bakija) on Friday, October 18, 2013 - 06:19 pm: Edit

Oh, absolutely--I don't think anyone is suggesting changing anything till the tournament is over.

By Ken Kazinski (Kjkazinski) on Friday, October 18, 2013 - 10:41 pm: Edit

Correct no changes to the ships until we finish CoN.

Once the tourney is done I would like to finalize the changes and get a good copy we can use.

By Ken Kazinski (Kjkazinski) on Saturday, October 19, 2013 - 11:51 pm: Edit

I received this from Robert Schirmer

Ok, here is what I have in terms of raw data – not much beyond the Council results, but perhaps Barry Kirk has more:

Winner Rating	Loser Rating	Winning Ship	Winning Ship	Losing Ship	Losing Player	Date YY/MM/DD / Source
1921	1644	MAE	BaldnForty	ZIN	GaiusFrackingBaltar	2011/03/25
1806	2037	MAE	Ron_Pilotte	ORI	Pimp_Daddy_Sackett	2013/10 Co5N
1806	2220	MAE	Ron_Pilotte	AND(PLAYTEST)	Indy	2013/10 Co5N
1922	2299	MAE	BaldnForty	ORI	AdmiralDZZI	2013/10 Co5N
1922	2641	MAE	BaldnForty	HYD	Moose	2013/10 Co5N
1922	2679	MAE	BaldnForty	KLI	Bakija	2013/10 Co5N
1922	1818	MAE	BaldnForty	RKR	Ben_Kalb	2013/10 Co5N
1922	2037	MAE	BaldnForty	ORI	Pimp_Daddy_Sackett	2013/10 Co5N
2202	1806	ISC	Agent_of_Change	MAE	Ron_Pilotte	2013/10 Co5N
2202	1806	ISC	Agent_of_Change	MAE	Ron_Pilotte	2013/10 Co5N
2037	1922	ORI	Pimp_Daddy_Sackett	MAE	BaldnForty	2013/10 Co5N
2521	1921	RFH	Lord_Goofy	MAE	BaldnForty	2011/03/14
1764	1922	RKE	Vandor	MAE	BaldnForty	2013/05/27
2395	1866	HYD	Tartan_Army	MAE	ScottishEngineer	2013/05/28

The games not marked Co5N are all NetKill results.

I did a quick analysis of the data per my SUPR approach. In outline, I assume (a) player ratings are as per my v5 SUPR data – see first two columns above; (b) assume a prior estimate of Maesron balance equivalent to one win and one loss between two equal players – the same as used in SUPR; (c) one overall Maesron balance score is calculated across all the data, without an attempt to estimate RPS vs. each opposing ship type, because there is too little data for that; and (d) Barry did not change the MAE SSD between the various matches above, which may not be the case.

Given the assumptions above, I get the curve in the attached picture. The x-axis values are hypothesized MAE win percentages for equal players, grouped in single percentage point bins. So, for example, 0.50 on the x-axis is the hypothesis of perfect balance, where the MAE would win 50% of its games between equal opponents (actually a bin going from 49.5% to 50.5% win probability). The y-axis is the probability that each x-axis hypothesis is true, given the match results above and the assumptions I list out. So, for example, the results show slightly less than a 1% chance that the 50% win hypothesis is true.

Anyway, a glance at the curve suggests that the MAE is likely too strong. To quantify it in summary terms, I divide the x-axis into three regions (too weak, balanced, too strong) and integrate the area under the curve in each region. Here’s what I get:

Case 1.
MAE Win Probability with Equal Players < 45% (Too Weak).
Probability = 2% Given Data

Case 2.
MAE Win Probability with Equal Players >= 45% and <= 55% (Balanced).
Probability = 10% Given Data

Case 3.
MAE Win Probability with Equal Players > 55% (Too Strong).
Probability = 88% Given Data

So that’s what I’ve got. Obviously there are a few caveats. You have to buy into my whole approach, for one, and also there’s a chance that everyone is using suboptimal tactics right now, given that the MAE is relatively new.

By Ken Kazinski (Kjkazinski) on Saturday, October 19, 2013 - 11:55 pm: Edit

After looking over the data it seems there really have not been that many matches outside of this years Council of Nations.

If anyone wants Roberts e-mail with the curve let me know.

By Robert W. Schirmer (Rwschirmer) on Sunday, October 20, 2013 - 12:23 pm: Edit

I just caught up with this thread and saw that the MAE was used in World League 2013. That adds the following games to the list above:

1922 2207 MAE BaldnForty HYD Duke 2013/10 WL
2490 1922 ORI Romwe MAE BaldnForty 2013/10 WL
2202 1922 WBS Agent_of_Change MAE BaldnForty 2013/10 WL
2731 1866 LDR Old_School MAE ScottishEngineer 2013/10 WL
2367 1866 RKR Spartan MAE ScottishEngineer 2013/10 WL

Repeating the balance estimate with the additional matches included gives:

Case 1.
MAE Win Probability with Equal Players < 45% (Too Weak).
Probability = 3% Given Data

Case 2.
MAE Win Probability with Equal Players ³ 45% and £ 55% (Balanced).
Probability = 16% Given Data

Case 3.
MAE Win Probability with Equal Players > 55% (Too Strong).
Probability = 81% Given Data

v/r

Robert

By Ken Kazinski (Kjkazinski) on Sunday, October 20, 2013 - 08:19 pm: Edit

Robert,

Based on this data you are expecting 81% of the time for the MAE to win 55% of the time, correct?

By Robert W. Schirmer (Rwschirmer) on Sunday, October 20, 2013 - 09:26 pm: Edit

Hi Ken,

There is an 81% chance that the MAE is too strong. Too strong is defined as the MAE being able to win more than 55% of its games between equal players, because its SSD is too good.

Again, this is just a quick cut where I made simplifications. One of the major factors not considered is if some or all of the players used suboptimal tactics due to a lack of familiarity with the MAE. Qualitatively, the MAE won 9 of 19 games, and the MAE players were generally not as strong as their opponents in terms of rating. That's where the result above is coming from, but of course it hinges on the ratings being about right for all these players in matches with the MAE.

Bottom line is I wouldn't put huge stock in the exact numbers until more data is available, but what's there now suggests the ship is on the strong side.

v/r

Robert

By Robert W. Schirmer (Rwschirmer) on Sunday, October 20, 2013 - 09:27 pm: Edit

Duplicate post deleted.

By Michael Kenyon (Mikek) on Monday, October 21, 2013 - 11:21 am: Edit

As best as I can tell, so far 2 people have managed to get a win in the MAE and if you add me, you get to the three of us that we have reported results from.

Would it behoove us to get results from more than 3 of us before we made any changes?

By Ken Kazinski (Kjkazinski) on Monday, October 21, 2013 - 06:57 pm: Edit

Michael,

There are 19 reports from Robert. It was the general consensus at Council of 5 Nations that the MAE was too good.

I think that this is probably the first time the ship has really been played by and against a large number of players at one time.

By Michael Kenyon (Mikek) on Tuesday, October 22, 2013 - 01:02 pm: Edit

Ken, I agree that there are 19 reports, but that represents reports coming in from 3 people flying the MAE.

Not sure what the standard is, but it seems to me to be early to make a move one way or the other on the ship.

As I understand the history of the tourney, ships have a tendency to either improve or fall down as people figure out the tactics for or against them. It seems excessive to say that a ship, any ship, is clearly above or below the line after a weekend of playing without people walking away, looking at it and figuring out how to defeat the dang thing.

By Ken Kazinski (Kjkazinski) on Tuesday, October 22, 2013 - 08:03 pm: Edit

One of the problems I see is that the tournament community will debate something until nothing gets done.

We have had no resolution on getting new ships into the tournament. I agree that we should not make hasty decisions but inaction is just as bad.

By Andrew J Koch (Droid) on Tuesday, October 22, 2013 - 09:09 pm: Edit

Funny I have had little interest in the Maesron but fought Barry's Maesron last night in the CO5N Semi, and so just read some of this thread.

We are not quite done with the fight and I am not sure right now when we will be. Suffice to say for now that I avoided R8 on T2 just as he describes up thread here, and I think that makes it quite tough on the Maesron. Maybe Barry can speak to it better than I can after we are done, but i think it saved me quite alot of internals when the smoke cleared after T3.

By Peter Bakija (Bakija) on Wednesday, October 23, 2013 - 08:49 am: Edit

I think very little is lost at this point by reducing the #1 shield to 30 boxes and pulling out 2 of the APRs (giving it the standard of 30-30-24-24 shields and 38 power, just like all the other ships that have some 0 energy weapons). Keep playing it like that. See if it takes a nose dive.

I'm not convinced that it is wildly over the top too good. And it seems likely that people just need experience playing against it. But erring on the side of caution and pushing it to standard shields and power seems completely reasonable at this point.

By Ken Kazinski (Kjkazinski) on Wednesday, October 23, 2013 - 08:52 pm: Edit

Based on the review at Co5N, Robert's analysis and the discussions here I believe the above changes represent a good change to make after Co5N tournament is closed.

By Andrew J Koch (Droid) on Wednesday, October 23, 2013 - 09:25 pm: Edit

I think it needs 40 power.

By Michael Kenyon (Mikek) on Wednesday, October 23, 2013 - 09:48 pm: Edit

I hold it's too early to tell. I believe that no matter what ship you add, or which of the existing ships you change, in which you're going to get skewed results before the true state pans out.

If we were to follow that logic, the current playtest andro is underpowered and needs to be improved, for example, and probably should have been changed 2-3 times by now.

By George Duffy (Sentinal) on Thursday, October 24, 2013 - 05:32 pm: Edit

I think the data is a bit eschewed in the fact that all of the Co5N data is based off of just one tournament. Going by the proposed logic Andy Koch's AuxBC could also have the same or even slightly higher results in comparison to the MAE. It fought the same number opponents with similar results.
Most of the data on the MAE is based off of Barry's play. Even if he has a lower overall rating, he has more practical experience flying that ship than all of his opponents have in facing off against it. Just because he placed higher in the tournament doesn't mean that the ship must be too strong because he did so well.

By Peter Bakija (Bakija) on Thursday, October 24, 2013 - 05:36 pm: Edit

I'm not gonna argue that the ship is necessarily too strong. I am gonna argue that the ship shouldn't have the two extra front shield boxes. And it should probably only have 38 power, given that it is similar to the Shark, which has 38 power, in that it has torps, power free drone weapons, similar internal fortitude (16 even hull, 12 phasers), and while it has 2 turn weapons, it can front or back load them for 1+X or X+1, giving it a significant edge, powerwise, on loading those guns.

By Ken Kazinski (Kjkazinski) on Thursday, October 24, 2013 - 07:09 pm: Edit

I agree with Peter on this. We had a long discussion about the MEA at Co5N. We could not find a good reason the MEA has 2 extra shield boxes and 2 extra power.