• The forum software have been upgraded to the latest version.

    If you notice anything that looks off, or does not work, please let us know.

    For more information, click here.

TOW Unit Tier List 2.0

R
Razordon

RipperDerek

Active Member
Messages
340
Likes Received
95
Trophy Points
28
So, I did a thing. I expanded my combat simulator into an auto-battler, and entered in a bunch of different lists and had them fight against each other.

The auto-battler abstracts out things like movement. When something tries to charge for example, it has a % chance to be successful, based on its movement range and the size of the playing field. This is also how things like range for shooting attacks or wizardly dispels work. Otherwise, I think it should be getting pretty much all the rules correct.

Not taking into account maneuvering and terrain means it is obviously not a perfect simulation, but the combat and shooting math should all be correct. It supports some 70 special rules I've had to program into it manually for things like Poison, Armor Bane, Unbreakable, etc. I have been using this on my own to test out different lists, but while doing so it occurred to me that you could use it to make a tier list. I did that and thought the results were interesting, so I decided to post them here.

What I did was to make 2000 point army lists composed of only one type of unit, and battle them against a gauntlet of Brettonia, Tomb King, Empire, and Ogre Kingdoms lists. Each type of unit was simulated for 1000 games. The following chart shows how many wins each unit type had out the 1000 games that were played.
  • 939 - Skinks w/Javelins
  • 841 - Skink Chiefs on Stegadons
  • 773 - Oldbloods on Carnos
  • 724 - Stegadons
  • 708 - Ancient Stegadons
  • 707 - Bastiladons
  • 604 - Saurus Cavalry
  • 486 - Temple Guard
  • 329 - Kroxigor
  • 316 - Saurus
  • 212 - Ripperdactyls
  • 192 - Scarvet Cowboys
  • 161 - Razordons
Each list had a general, and then the rest of the army was filled in with duplicate copies of the unit to be tested. The top performing list was 380 skinks, with almost a 94% win rate. It's also one of the only lists that is actually legal. Also, the difference between Stegadons, Ancient Stegadons, and Bastiladons is almost certainly within whatever the margin of error is, so this data says those three things simmed about the same.

Anyways, I hope this endeavor proves helpful to people other than just me. Good luck, and happy lizarding.
 
RipperDerek said:
So, I did a thing. I expanded my combat simulator into an auto-battler, and entered in a bunch of different lists and had them fight against each other.

The auto-battler abstracts out things like movement. When something tries to charge for example, it has a % chance to be successful, based on its movement range and the size of the playing field. This is also how things like range for shooting attacks or wizardly dispels work. Otherwise, I think it should be getting pretty much all the rules correct.

Not taking into account maneuvering and terrain means it is obviously not a perfect simulation, but the combat and shooting math should all be correct. It supports some 70 special rules I've had to program into it manually for things like Poison, Armor Bane, Unbreakable, etc. I have been using this on my own to test out different lists, but while doing so it occurred to me that you could use it to make a tier list. I did that and thought the results were interesting, so I decided to post them here.

What I did was to make 2000 point army lists composed of only one type of unit, and battle them against a gauntlet of Brettonia, Tomb King, Empire, and Ogre Kingdoms lists. Each type of unit was simulated for 1000 games. The following chart shows how many wins each unit type had out the 1000 games that were played.
  • 939 - Skinks w/Javelins
  • 841 - Skink Chiefs on Stegadons
  • 773 - Oldbloods on Carnos
  • 724 - Stegadons
  • 708 - Ancient Stegadons
  • 707 - Bastiladons
  • 604 - Saurus Cavalry
  • 486 - Temple Guard
  • 329 - Kroxigor
  • 316 - Saurus
  • 212 - Ripperdactyls
  • 192 - Scarvet Cowboys
  • 161 - Razordons
Each list had a general, and then the rest of the army was filled in with duplicate copies of the unit to be tested. The top performing list was 380 skinks, with almost a 94% win rate. It's also one of the only lists that is actually legal. Also, the difference between Stegadons, Ancient Stegadons, and Bastiladons is almost certainly within whatever the margin of error is, so this data says those three things simmed about the same.

Anyways, I hope this endeavor proves helpful to people other than just me. Good luck, and happy lizarding.
Click to expand...
Very cool!
 
Interesting....i don't know if this is the best way to testing but why not.

A bit surprised on this
  • 724 - Stegadons
  • 708 - Ancient Stegadons

But i guess it depends by both initiative 1 and higher points cost. Do your programme calculate also ranged units or magi? It would be interesting to se with engine of gods and also salamanders.

Another thing....no ripperdactyls are NOT so bad. In nearly everything else I can agree (more or less)
 
Yes, it simulates both ranged attacks and magic. Although magic is somewhat weird because spells that do things like let you move twice or block line of sight just work by changing the range %s around.
 
Interesting stuff!

A couple of notes based on my games so far:
Saurus Cavalry seem to fare a bit better in simulation than I thought they would.

Bastiladon looking better and better every day. I was initially feeling pretty meh about them but after a couple of SOLID games with them I'm starting to see what the laser lizard can do. Great to see stats backing it up.
 
My favourite the Troglodon? I don't even want to know...
 
I had the chance to believe and try salamanders in the last 2 games. Against T3 infantery they are good but i played one to maximize mobility and could easily flee beside/behind an infantery block if charged). The only problem is to rally them with Ld 5 if the general is not nearby.
However is pretty useless against cavalry/monsters/some skirmisher. I don't know if it worth to play them but it is probably worth against infantery lists or in tournaments with lot of infantery due restrictions.
 
Kalisto said:
I had the chance to believe and try salamanders in the last 2 games. Against T3 infantery they are good but i played one to maximize mobility and could easily flee beside/behind an infantery block if charged). The only problem is to rally them with Ld 5 if the general is not nearby.
However is pretty useless against cavalry/monsters/some skirmisher. I don't know if it worth to play them but it is probably worth against infantery lists or in tournaments with lot of infantery due restrictions.
Click to expand...

I don't think RAW they even can use general LD. Salamanders are warbeasts and their troop type prevents them from it (no reroll from BSB either).
 
Rimbo said:
I don't think RAW they even can use general LD. Salamanders are warbeasts and their troop type prevents them from it (no reroll from BSB either).
Click to expand...
Oh didnt know that have to check it!

Edit:
Ok i agree war beasts cannot, but since the unit is mixed and you have skinks in them i would argue that they can as long as skinks are in the unit.
 
Kalisto said:
Ok i agree war beasts cannot, but since the unit is mixed and you have skinks in them i would argue that they can as long as skinks are in the unit.
Click to expand...

Tricky, since undisciplined rule says that "A unit with this rule cannot use the Inspiring Presence"

Skinks don't have undisciplined but they're still part of the same unit.
 
Mmm that's true. Probably they cannot but it's a bit strange since salamander is a mixed unit =(
Probably it's an exception that is not complied by the actual rules (so tired of such exceptions...)
Before neither a hero could join salamanders but now isn't forbidden either right? Just thinking if a skink hero with Ld6 could help a bit =) (probably not worth it but just theoretically)
 
Last edited:
BeardedLizard said:
Interesting stuff!

A couple of notes based on my games so far:
Saurus Cavalry seem to fare a bit better in simulation than I thought they would.

Bastiladon looking better and better every day. I was initially feeling pretty meh about them but after a couple of SOLID games with them I'm starting to see what the laser lizard can do. Great to see stats backing it up.
Click to expand...
Keep in mind this simulation is a list of only Saurus Cav, so when one unit fails to charge due to stupidity, chances are another doesn't

Where as in most of our lists we only have 1 unit of saurus cav. So you you will always notice when stupidity hits.
 
Isn fairness to the Troglodon, it's a support unit

This would be like saying a list of only techmarines does horrible

Yeah. Because the value in the techmarine isn't what he does himself. It's what he does for other units

Same for a Troglodon iMO

Now that doesn't mean the trog isn't underwhelming. But I don't think the trog is as dismal as that score implies
 
This approach does show how the units match up against average enemy units in general, but it doesn't account for synergies between our units. It's definitely possible that certain combinations of units could perform better, if they cover each other's weaknesses, etc. But the number of possible combinations gets huge quickly.
 
You must log in or register to reply here.
Back
Top