Jump to content
punkUser

X-Wing Probability Calculator - Second Edition!

Recommended Posts

Memorizing dice statistics is fun! Everything about this tool is great. Super thanks

I have already confirmed that Norra Wexley's ability is really good. Also, reinforce remains strong. Palpatine defended by reinforce in 2e sounds scary.

Next I'll have to see how focus compares with multiple calculate tokens in different circumstances.

Share this post


Link to post
Share on other sites

New update with a bunch of improvements:

  • Changed the way the token graph is displayed. It now shows expected (positive) token counts after the attack(s) for both attacker and defender rather than the delta with their initial tokens as before. This allows us to pack more information into a smaller chart and allows us to show effects like the chance of becoming ionized more clearly. Ideally this is also more intuitive to people once we adapt to the new display. It will admittedly feel a bit confusing for those of us used to the old chart for a little bit though.
  • Improved defender token spending logic in a few situations (notably when reinforced) due to using a search for defense now as well. If you notice any issues with this (such as defender spending being worse than before or expected) please comment here as it is a fairly large change to the logic.
  • Added Shara Bey on defense
  • Added Gavin Darklighter
  • Added Ion weapons. These will show up on the expected token chart as "Ionized (1)", "Ionized (2)" and so on as applicable. The idea is that those values show the chance to land that many ion tokens on the target, which is usually more useful than knowing the average amount of tokens. For instance, if the defender starts with zero ion tokens, "Ionized (1/2/3)" is the chance to ionize a small/medium/large base respectively.
  • Added Zeb pilot

As always let us know if you notice anything that seems off. Enjoy!

Share this post


Link to post
Share on other sites

Added a few more things today:

  • Added a "how to modify defense dice" page as well. While fewer reroll effects make this less necessary than attack there's still a few interesting situations.
  • Improved some of the output on the "how to modify attack" pages to include calling out certain effects like changing dice and such. There's probably a good middle-ground between spelling everything out by name and saying nothing and I think we're closer to it now than before, but let me know if it seems intuitive enough now.
  • Added Boba Fett rerolls (just 1 for now, but will add 2 and 3 as options in the future as well).

As always let us know if you find any issues or see anything unexpected.

Share this post


Link to post
Share on other sites

More updates!

  • Made Boba/Horton rerolls generic pilot skills (reroll up to 1/2/3)
  • Added Saw Gerrera Pilot (1 reroll)
  • Added L3-37. This one is a bit of a tricky card with some non-trivial optimal usage and thus it is the first one to use a search while modifying the opponents dice. Has some interesting behavior for you guys to explore :)
  • A variety of other cleanups, tweaks and fixes

As always please let us know if you run into any issues, particularly with L3-37 as she exercises some new code paths. Enjoy!

Share this post


Link to post
Share on other sites

A few more things pushed live:

  • Added Scum Lando crew... this one is fairly tricky as there are a lot of options on how to use it, so it actually does a bit of a search just to decide if and which dice to reroll, and using which green token. Give it a shot and let me know if you see it doing anything unexpected.
  • Added Elusive
  • Added Feroph and Vermeil
  • Added Serissu
  • Added Sabine lancer pilot
  • Added RAC
  • Added Agent Kallus

Enjoy! Note that we'll be giving some thought into how to reorganize the UI a bit (particularly for multi-attack) in the future as the list of effects grows.

Share this post


Link to post
Share on other sites
On 6/14/2018 at 3:49 PM, punkUser said:

For example - if I have a lock and a calculate token and roll 2 blanks and a focus, which dice do I reroll? (Answer: all three!)

Haha you just blew my mind. I had not thought about this at all, glad you guys are crunching the numbers for us. 

Share this post


Link to post
Share on other sites

A few more effects for people theory-crafting their new 2.0 lists:

As usual, give it a shot and let me know if you run into any bugs or suspicious behavior!

Share this post


Link to post
Share on other sites

A few resistence and first order things from the preview articles:

  • Added Finn gunner
  • Added Heroic
  • Added Fanatical

Heroic seems like a pretty minor effect which makes sense since it's only triggered by entirely blanking out. Might have some interesting effects on the outliers though so play with it a bit and see if it does anything cool.

Enjoy!

Edited by punkUser

Share this post


Link to post
Share on other sites

Alright I just pushed a moderately sized update with a new calculator for people to play with: Ship Durability!

There's some info in the FAQ page about this calculator but the main purpose is to get a rough idea of how many shots a given ship can take before being destroyed. There's a dropdown to pick the nature of the attacker(s) (i.e. naked 2 dice vs. double-modded 4 dice are two very different situations and different defenders perform differently as you can now see easily) and then it simulates and compares the relative number of shots your ship can take to a variety of other ships in the game. Like the other simulations, it does this correctly and exhaustively by following the full set of states and dropping branches in which the ship is destroyed rather than approximating things with expected damage or so on which misses correlated factors, including things like Stealth Device, Iden, etc.

As per the FAQ, this is meant to give a general idea of relative ship durability, and thus there's no defender tokens, varied ranges or anything like that as that would introduce further questions around how many shots are being taken in a given turn, actions and tokens and each turn and so on. If you're interested in specific cases the regular calculator allows arbitrary options to be set and is more appropriate.

Also note: crits are not yet being modeled in terms of tracking stuff like Directs, Fuel Leak, etc. This will come in a future update, at which point the relative ratio of shields to hull will matter.

There were a few other small updates since the last post as well, but that's the big one. Give it a try and let us know what you think!

http://xwing.gateofstorms.net/2/ship_durability/

Share this post


Link to post
Share on other sites

Quick update:

  • Added a 2.0 version of the simpler "presets" calculator for multiple attacks: http://xwing.gateofstorms.net/2/multi_preset/. While this can be accomplished with the main 2.0 calculator of course, it's often easier for simple cases to use the dropdowns. In the future will have a link that lets you jump from the simple one of the equivalent settings in the full calculator as well.
  • The ship durability calculator now uses the preset form as well for consistency.
  • Added a few more "reference" ships to the durability results.

Will be adding more preset options in the future as well, similar to the 1.0 multiple attacks calculator.

NOTE: Now that 2.0 is almost out and most of the functionality from the 1.0 calculator exists (in expanded forms!) in the 2.0 version, the 1.0 version will likely be removed in a future update for UI and code iteration simplicity. If this is a huge problem for you please let me know here or via GitHub, along with how long you anticipate wanting it to stick around. Note that it's of course always possible to sync and continue to use and modify the 1.0 stuff as long as you want via the source code, I just don't want to continue to maintain it forever.

Share this post


Link to post
Share on other sites
19 hours ago, Admiral Deathrain said:

Meh, so 3PO on Leebo is always worse than keeping the token, that is a bummer and calls for a serious price reduction on what is almost always a worse perceptive co-pilot. 

Not sure what you mean by "always worse". There are obviously a fair number of cases where it's quite a bit better. Picking an arbitrary but reasonable case at random, 3 attackers all with focus:

No 3P0: http://xwing.gateofstorms.net/2/multi_preset/?d=AggAAAEAAAA&a1=AgM&a2=AgM&a3=AgM - Expected 3.233 damage
W/ 3PO: http://xwing.gateofstorms.net/2/multi_preset/?d=AggAAgEAAAA&a1=AgM&a2=AgM&a3=AgM - Expected 2.666 damage

The calculator's use of 3P0 isn't completely optimal either, although it is a reasonable heuristic. If you had more knowledge of which attacks you expect to be incoming over a turn you can do slightly better on average though.

In most cases Leebo effectively has a soft glitterstim with no stress. 3P0's effect certainly helps in cases where you expect relatively accurate/large numbers of hit attacks coming in, and pairs well with the fact that you get two calculate tokens via him (which itself is quite strong).

Edited by punkUser

Share this post


Link to post
Share on other sites

Can I piggyback on this thread being bumped to make a request based on something I ran into this morning?

Could you add Wedge's ability to the attacking pilot ability drop down, and Crack Shot to the talent options?

I know that for both abilities on just one attack, you could ordinarily just drop the defence dice count. But when you're trying to model the expected outcome of multiple attacks on one defender, it's difficult to account for both of these effects when they only apply to one of the attacks. 

The tool's really great, by the way. I'm amazed how often I go back to it after finding it, it's so good for working out where to focus on list building - whether certain upgrades or worth it over just choosing to fly a certain way etc. I love all the stats for expected hits and expected remaining tokens on an attack by attack basis.

Thanks so much for all your work on it!

Share this post


Link to post
Share on other sites
20 minutes ago, GuacCousteau said:

Can I piggyback on this thread being bumped to make a request based on something I ran into this morning?

Of course, that's the primary purpose of the thread! Feel free to post requests here any time.

Quote

Could you add Wedge's ability to the attacking pilot ability drop down, and Crack Shot to the talent options?

Wedge's ability is effectively already there via the "+/- defense dice" thing below the attacker dice. Just set that to -1. Indeed it's for precisely the purpose you describe: controlling defense dice modifiers like range and Wedge and outmaneuver and so on on a per-attacker basis. We could add Wedge to the pilots list for simplicity but it would just duplicate that functionality.

Crack shot is on the TODO list. It has a bunch of non-trivial interactions (ex. with reinforce) that we were waiting to sort out. The mechanics of it are now known, it just has some fiddly details around defender token spending in cases that can't really currently happen in the real game, but have to be handled in the simulator of course. It's definitely still near the top of the TODO list though - we want it as well :)

Quote

The tool's really great, by the way. I'm amazed how often I go back to it after finding it, it's so good for working out where to focus on list building - whether certain upgrades or worth it over just choosing to fly a certain way etc. I love all the stats for expected hits and expected remaining tokens on an attack by attack basis.

Thanks, glad to hear it's useful! In addition to squad building I often use it after a game where I'm wondering if I made a poor decision or if some outcome was just lucky dice or similar. It's great for separating out those "oh ok that really was a 1% case" from places where you did something that was actually riskier than you thought.

Edited by punkUser

Share this post


Link to post
Share on other sites
5 minutes ago, punkUser said:

Wedge's ability is effectively already there via the "+/- defense dice" thing below the attacker dice. Just set that to -1. Indeed it's for precisely the purpose you describe: controlling defense dice modifiers like range and Wedge and outmaneuver and so on on a per-attacker basis. We could add Wedge to the pilots list for simplicity but it would just duplicate that functionality.

I'm an idiot. 

I've been using the tool a bunch and I never once noticed the +/- defence dice option. Request retracted ?

I'm booking an eye test tomorrow...

6 minutes ago, punkUser said:

Crack shot is on the TODO list. It has a bunch of non-trivial interactions (ex. with reinforce) that we were waiting to sort out. The mechanics of it are now known, it just has some fiddly details around attacker token spending in cases that can't really currently happen in the real game, but have to be handled in the simulator of course. It's definitely still near the top of the TODO list though - we want it as well

Gotcha, makes total sense.

Now that I know about the defence dice modifier, I can just use that to simulate expected hits at least. It'll mess up the expected token counts a little, but I can deal with that. Good to know you're looking at getting it in eventually though. 

10 minutes ago, punkUser said:

Thanks, glad to hear it's useful! In addition to squad building I often use it after a game where I'm wondering if I made a poor decision or if some outcome was just lucky dice or similar. It's great for separating out those "oh ok that really was a 1% case" from places where you did something that was actually riskier than you thought.

I'll definitely bear that in mind. I've for sure had those moments; so many little things you end dissecting in your head for hours after a close game. 

Share this post


Link to post
Share on other sites

Quick update:

  • By request, added a 7th attack option to the forms
  • Added some common preset links to the dropdown
  • Removed 1.0 calculator. 2.0 for life! :)

More stuff in the pipe coming soon, but wasn't quite ready for prime time yet.

Share this post


Link to post
Share on other sites

ugh this again.  this mentality is completely flawed. its a game which involves decision making, which is not computable within your framework. the math you're using ignores the most important element of the game (movement).  it isnt a complete computation, so its pointless. ironically the only triumph of math wing was convincing a lot of people to play min max garbage like triple jumps (which most people credit with ruining the game mechanically)

but that leads to the second point, which is the goal of mathwing is to help people objectively know that certain ships are better than others for the purpose of power gaming.  obviously people will know whats good and choose it, but the idea of mathwing is basically poor sportsmanship. its the same crap that turned local sports teams into pay to win mercenary troops. its a game, just make youre own list and play it for fun

Share this post


Link to post
Share on other sites
4 hours ago, Vontoothskie said:

ugh this again.  this mentality is completely flawed. its a game which involves decision making, which is not computable within your framework. the math you're using ignores the most important element of the game (movement).  it isnt a complete computation, so its pointless. ironically the only triumph of math wing was convincing a lot of people to play min max garbage like triple jumps (which most people credit with ruining the game mechanically)

You might have the wrong thread dude... this isn't "mathwing" per se and makes no attempts to categorize which ships are better than others. If you read the About page of the calculator it makes it extremely clear that the point is to allow comparing various *specific* situations and upgrades and so on to help *inform* list building and decision making. We're not trying to generalize anything, we're just trying to provide a tool and help guide people as to how to do better analysis. Indeed at every possible place where there's a chance of providing over-generalized or ambiguous data in the tool you might notice that we have explicitly forced people to zone in on a deterministic and specific result so they know exactly what they are getting.

Fundamentally this is just basic dice probabilities with a UI specialized with the X-Wing upgrade names. It's *not* mathwing or any similar attempt to model ship efficiency, costs or otherwise.

In any case feel free to start a separate thread if you want to have a different discussion; the main point of this thread is to inform folks on updates and usage of the tool.

Edited by punkUser

Share this post


Link to post
Share on other sites

Implemented a few more effects from the next wave:

  • Added "Hate", along with a new "max force" control. The latter is required for any abilities that regen force during an attack such as Hate or Luke Skywalker. Thus there's a minor bit of incompatibility where Luke effectively had it silently set to "2" in the past, but now it needs to be explicit set, so remember to do that if using any of these effects!
  • Added Predictive Shot. Remember to add a force point if you want it to be used.
  • Added Advanced Optics.
  • Some misc UI cleanup/movement related to the new stuff.

Enjoy and let me know if you see any issues! I haven't had time to do extensive testing on the new stuff beyond the initial "it seems to be working properly" so let me know if you see anything funky.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×