47 replies to this topic

[LordNothing]

not supprised about the lights. the durable ones are a little better off for lower skill light pilots. i do better in the worst 2 in the class. i also find the worst medium, the ice ferret, is rather high on my mech performance spreadsheets. in fact there are a lot of mechs in the 'worst' categories that i find to be really good mechs. exe can be brutal if you can exploit its mobility, its a very useful thing to have in the nascar rich environment that is quick play. sunspider is also one ive been doing fairly well with, especially the quad ballistic with a pair of uac2s and a pair of uac5s.

also can haz graphs?

Edited by LordNothing, 21 January 2019 - 06:58 PM.

[Mr Andersson]

@OP

I appreciate that you have taken the time to put this together, but there are two things that diminish the value of the data presented.

Firstly, as you have stated, you are tier 3. That means the matches you dropped in could contain any player in the MWO player base. They could be elite comp level players or complete potatoes. We don't know. Therefor, it is very hard to judge the value of the numbers.

Secondly, as any statistician would tell you, 100 is not enough of a sample size to draw any form of conclusions from.

[justcallme A S H]

GeminiWolf, on 21 January 2019 - 06:17 PM, said:

Did you take into account the ping of each mech as well? that effects pilot effectiveness.

Rubbish. I play with between 230-400ms ping. Be that on US/EU servers, QP/GQ/Solaris - It does not really have any major effect.

GeminiWolf, on 21 January 2019 - 06:17 PM, said:

This study is flawed.

So was your post, so what?

[GeminiWolf]

justcallme A S H, on 21 January 2019 - 08:51 PM, said:

Rubbish. I play with between 230-400ms ping. Be that on US/EU servers, QP/GQ/Solaris - It does not really have any major effect.





So was your post, so what?

A S H, you are a great pilot so your ping would not effect you, however most other players it has an effect on. So taking that into account would not give good numbers on specific mechs. I do mean GREAT pilot. So again the statistics are flawed.

[Acersecomic]

Darian DelFord, on 21 January 2019 - 10:16 AM, said:

Sheesh I have not seen a firestarter in months of play. Most common lights are Piranha, Wolfhound, urbie Locust and Flea on the IS side

I actually started using it. 4 flamers in torso and 4 medium pulses in arms, the other build is 8 medium lasers and jumpjets.

[justcallme A S H]

GeminiWolf, on 21 January 2019 - 09:25 PM, said:

A S H, you are a great pilot so your ping would not effect you, however most other players it has an effect on. So taking that into account would not give good numbers on specific mechs. I do mean GREAT pilot. So again the statistics are flawed.

250ms ping is 250ms ping. If I have it and someone else has it - it becomes irrelevant and then just down to raw effectiveness.

Anyone using Ping as a basis to form an argument upon when it comes to MWO, that's rocky at best. There are games out there that it obvious does make quite a bit of difference but not here in MWO as HSR does fix quite a lot of that up.

The only time it does is when someone is dropping packets and essentially jerks across the screen / teleports 5-10m at a time etc. That's totally different though.

[Knight Captain Morgan]

LordNothing, on 21 January 2019 - 06:57 PM, said:

not supprised about the lights. the durable ones are a little better off for lower skill light pilots. i do better in the worst 2 in the class. i also find the worst medium, the ice ferret, is rather high on my mech performance spreadsheets. in fact there are a lot of mechs in the 'worst' categories that i find to be really good mechs. exe can be brutal if you can exploit its mobility, its a very useful thing to have in the nascar rich environment that is quick play. sunspider is also one ive been doing fairly well with, especially the quad ballistic with a pair of uac2s and a pair of uac5s.

also can haz graphs?

Yeah I found it amusing that the most popular mech had the lowest average match score and the least popular had the highest. I don't have a spreadsheet graph program but if someone wants to reformat this info into something easier on the eyes, feel free Or better yet, attempt to reproduce the results with better methodology etc. More data points the merrier. I just ask that you use a mech other than the Shadow Cat as I'd enjoy seeing some unbiased data on that chassis.

Edited by Andres Gomez, 22 January 2019 - 04:19 AM.

[RickySpanish]

Mr Andersson, on 21 January 2019 - 07:21 PM, said:

@OP

I appreciate that you have taken the time to put this together, but there are two things that diminish the value of the data presented.

Firstly, as you have stated, you are tier 3. That means the matches you dropped in could contain any player in the MWO player base. They could be elite comp level players or complete potatoes. We don't know. Therefor, it is very hard to judge the value of the numbers.

Secondly, as any statistician would tell you, 100 is not enough of a sample size to draw any form of conclusions from.

Surely T3 would be a great place to sample from, since it includes the whole spectrum of MWO players?

[Bigbacon]

Darian DelFord, on 21 January 2019 - 10:16 AM, said:

Sheesh I have not seen a firestarter in months of play. Most common lights are Piranha, Wolfhound, urbie Locust and Flea on the IS side

because they are such garbage today. Slow, sluggish, damage pinatas for anyone. Miss the firestarter.

[Flyby215]

Wow! That's pretty neat info.

I'm scanning through the data to find mechs with, to my opinion, have obvious loadouts. The Dragon for example is likely running 3xUAC2s, the Awesomes, Supernovas, and Stalkers are likely in missile boat configurations, and the Huntsmen in an SRM configuration. All allow for high damage regardless of win/loss.

That said, the Awesome has a combination of high damage and high win-loss, translating into good match scores. The Supernovas and Stalkers might be assumed to have good damage but poor win-loss translating into average match scores.

This makes me wonder about Warhawks, with a 0-7 win/loss but an average match score of 211. Similarly Catapults with a relatively good win/loss having a relatively average 216 average match score.

Entertaining to think about. Really surprised to see that many Catapults, Mad Dogs, and Locusts. I hardly see any of them anymore.

Edit: Fixing data.

Edited by Flyby215, 22 January 2019 - 07:23 AM.

[Knight Captain Morgan]

Flyby215, on 22 January 2019 - 07:21 AM, said:

Wow! That's pretty neat info.

I'm scanning through the data to find mechs with, to my opinion, have obvious loadouts. The Dragon for example is likely running 3xUAC2s, the Awesomes, Supernovas, and Stalkers are likely in missile boat configurations, and the Huntsmen in an SRM configuration. All allow for high damage regardless of win/loss.

That said, the Awesome has a combination of high damage and high win-loss, translating into good match scores. The Supernovas and Stalkers might be assumed to have good damage but poor win-loss translating into average match scores.

This makes me wonder about Warhawks, with a 0-7 win/loss but an average match score of 211. Similarly Catapults with a relatively good win/loss having a relatively average 216 average match score.

Entertaining to think about. Really surprised to see that many Catapults, Mad Dogs, and Locusts. I hardly see any of them anymore.

Edit: Fixing data.

The Warhawk is one of my absolute favorites in TableTop. I was disappointed to see it go winless. It even had a worse ratio than the disconnects.

[Mr Andersson]

RickySpanish, on 22 January 2019 - 05:15 AM, said:

Surely T3 would be a great place to sample from, since it includes the whole spectrum of MWO players?

How to explain?

Imagine you want to evaluate the dominant strategies in [insert favorite team sport here]. Would you then look at games being played by the entire population, from 8 year olds to 80 year olds, and from every level of skill? Or would you just look at the professional leagues?

[TheCaptainJZ]

The OP was simply data--it wasn't looking for the best mechs, just the most popular.

[Prototelis]

I would be interested in knowing how many of those PIR pilots you saw more than once.

[Mr Andersson]

TheCaptainJZ, on 22 January 2019 - 08:55 AM, said:

The OP was simply data--it wasn't looking for the best mechs, just the most popular.

He lists average match score and win/loss ratio, doesn't he?

[Dimento Graven]

Mr Andersson, on 22 January 2019 - 08:51 AM, said:

How to explain?

Imagine you want to evaluate the dominant strategies in [insert favorite team sport here]. Would you then look at games being played by the entire population, from 8 year olds to 80 year olds, and from every level of skill? Or would you just look at the professional leagues?

True, but also when say the NFL is looking to the draft, they don't ONLY review players from the Big 10, or Ivy League schools, they pretty much recruit from all the NCCA associated institutions (maybe even from out side, though I'm not 100% sure of that, but seem to recall one or two player in the past).

This is good "general knowledge" info to have, and I think is a really great sampling of what the 'average' player will come across in T3.

It's why I'd be interested to see how it changes when he makes it to T2, and ultimately T1.

[Angel of Annihilation]

Hate to break it to the OP but the data is kind of useless because we had several events requiring specific mechs to be played to earn the rewards.

For example, I can't recall that last time I saw a IS Jenner in a match but with the Founders event going on, there was usually 1-3 Founders Jenners running around in about each match I played. This goes for the Founders Hunchback, Catapult and Atlas, all much more common in game due to the event. Same for all the Phoenix mechs. Tons of Phoenix Thunderbolt. Rarely see a T-bolt anymore but tons and tons of the Phoenix variant out and about during the event. Honestly tons of all types of Phoenix variants in the game that normally don't get played.

So yeah, if most of the data was collected while the events were going on, then the numbers aren't even close to accurate.

[Dimento Graven]

Angel of Annihilation, on 22 January 2019 - 09:47 AM, said:

Hate to break it to the OP but the data is kind of useless because we had several events requiring specific mechs to be played to earn the rewards.

For example, I can't recall that last time I saw a IS Jenner in a match but with the Founders event going on, there was usually 1-3 Founders Jenners running around in about each match I played. This goes for the Founders Hunchback, Catapult and Atlas, all much more common in game due to the event. Same for all the Phoenix mechs. Tons of Phoenix Thunderbolt. Rarely see a T-bolt anymore but tons and tons of the Phoenix variant out and about during the event. Honestly tons of all types of Phoenix variants in the game that normally don't get played.

So yeah, if most of the data was collected while the events were going on, then the numbers aren't even close to accurate.

Then you can ignore those 'mechs and look at what was brought that was not associated with the events and consider those.

Even though a significant presence of Founders or Phoenix games actually does give us some information about the game that we could really only assume prior this point: There's still a significant number of people who have been playing this game a VERY long time.

[Knight Captain Morgan]

Dimento Graven, on 22 January 2019 - 09:34 AM, said:

True, but also when say the NFL is looking to the draft, they don't ONLY review players from the Big 10, or Ivy League schools, they pretty much recruit from all the NCCA associated institutions (maybe even from out side, though I'm not 100% sure of that, but seem to recall one or two player in the past).

This is good "general knowledge" info to have, and I think is a really great sampling of what the 'average' player will come across in T3.

It's why I'd be interested to see how it changes when he makes it to T2, and ultimately T1.

As mediocre as I am, I wouldn’t be surprised if September 10th gets here before I tier up. But someone who is already tier 1 could collect their own data for comparison. Wink wink.

[DeskJockey]

Andres Gomez, on 21 January 2019 - 03:08 PM, said:

Yep, it was formatted with spaces. My main file wouldn't be much better as it is a .txt file created with notepad. I extracted the data from the screenshots manually using graph paper and a pen, then crunched the numbers with a calculator, then made a .txt file from which I copied/pasted into this forum post. I don't have a spreadsheet program so I just made due the old fashioned way.

So, Like, you did some actual math here? That is dedication!