Faction Ratings

Topic has gone off topic, but oh well. So I'm going to weigh in with my opinion here, as someone who plays Shaltari, and PHR, has corvettes in both, and amy's. And has played against a whole lot of factions. I'm not going to get in to the "who's played more games" argument, but rather, I'm going to provide some more maths, and a little bit of suggestions:

I am going to make some assumptions to make my points easier, We all understand what a R-Squared value is, and how it correlates to average rolls and that everyone's dice rolls differently, so your perceived average may be higher or lower, I've experienced both god tier rolling, and games where I get nothing above 3.

First the suggestions, You're mis-comparing and mis-using the two ships, as this has become a running comparison of the djinn and the amy's with the odd "look at corvette x' here and there. First things first, the Djinn, Scald, SCALD SCALD. Have I made my point? probably, have I elaborated, well lets do that now. The Djinn has scald, which means that it's capable of 'hitting up' or as we tend to call it "Uppercutting" Its job is to kill something bigger than it, and probably die doing it. Most of the scourge players I've played with/against use the Djinn to finish off/start/cripple big, high priority targets often ones with huge armor saves, EG, Battleships. For those who want to argue about PD and such, I'm assuming they are working in groups of 4, 4d6+8, Yeah, PD my rectum.
Amy's however, are built with 2 jobs in mind. Eliminating key assets (Strike carriers, fleet carriers ect) and waving a big middle finger at Jakarta's. I don't have to argue about PD, but If I took a group of 4 Amy's (firstly it'd be more expensive than the Djinn's and secondly it'd take up more slots) I would expect to be able to kill 1 key target, and maybe put the hurt on 1 or 2 others. the other thing to note, is that unlike Djinn's they don't have a 'back up' weapon or 'closing weapon' that can still be useful. Comparing them to go 'look at the numbers' is all well and good, but take a moment to consider what you are using them for.

Right, Next thing People keep throwing around averages, and other people keep going "yeah well your average is wrong because I can do incidental tests" This is a logical fallacy on BOTH sides. The first is the people going 'look at my average' this is called the fallacy of averages, The fallacy of the average is based on the false notion that the effect of a thing averaged out on a large scale is equivalent to an effect of the same thing on a small scale. A good example of this is this: Should rain start to fall in the atlantic ocean, where each drop falls will have no effect, However, what if a single drop of rain was to fall in to the mains circuit of the local radio control system for a airport? While yes, averages are a fantastic tool to compare and contrast things over a long term, One cannot discount the level of significance that comes from the chance of variation from the average. Ok, to those of you going "Your average is wrong because I can do Incidental tests" The fallacy you are performing is called "black Swan, White crow" you are going "ah ha! because I have found 1 black swan that means you're suggestion that most swans are white is wrong" Or equally "because I have bread 12 white crows Your supposition that on average crows are black is a lie!"
When you are judging ships on levels of power, You need to judge them on not what is the 'average' but on what the COULD do.
I'm going to break each ship down as best I can to compare them and score each ship individually, then I will compare damages with r-squared calculated, To save myself the effort of re-calculating a lot of things, I am going to ignore point defence and armor, because those two variables are way to high. you could have everything from 0 PD up to (highest in my games ever seen) 40. I am only calculating 'potential damage output' rather than 'actual damage taken' because once you know the potential output, you can calculate the 'damage taken' against anything you want.
I will also not judge them in a group, Only on a 1 to 1 basis to save myself time.
Also, The scoring is not going to 'value' each advantage, this is a Empirical comparison,
Let us begin,
Scan: Amy, 12, Djinn 6. Amy has a higher scan, 1 point to Amy
Sig: Amy, 2/12, Djinn 3, I have to give this one to the Amy, as on approach a low sig is useful.
Thrust: 12, and 12, Draw.
Hull, 4 and 4. Draw once again.
Armor: 5/4 and 5, I'm going to give this one ot the Djinn because to get that 4, the Amy has to take a huge sig boost.
PD, 6 for the Amy, and 4 for the Djinn. Amy gets a point.
Gropuing, Amy gets 1-2, Djinn gets 2-6, Djinn wins this, hands down.
3 'points' to the Amy, 2 'points' to the Djinn. at least for now. Weapons we are doing differently.
Lets start with the Amy. 1d3+1, at 3+ dealing 1 damage per attack. That will average out at about 3 attacks, with 2 hitting, for 2 damage. Now, That is your 'average' To account for statistical anomalies (or as our Black swan/White Crow friend would call them 'real rolls') we can take a RNG and tell it to generate for me, 400,000 dice rolls of a 1d3+1, Plot them on a graph, with a average of 3, and then ask 'what is the R-squared of this graph' The short answer is, high (somewhere about 0.8). You're maximum variation on EACH INDIVIDUAL amy is very low. I won't do it for hits, mostly because what we're looking at here is the percentage chance of an amy rolling 4 attacks vs it rolling 2 attacks and how much that impacts the gameplay over a huge amount of rolls.
So, to those of you going 'but mah real rolls' What this says is: for volume of attacks Amethysts are low, but reliable.

doing the same to Djinn d6+2 doing 1 damage each, with scald at lock 3. we can ignore scald. so, we get the PC to roll 400,000 d6+2, and we say our average is 5, We get a WAY lower R-squared (about 0.6 or so), so on 'average' we output more damage, But it is a more swingy weapon. You are less likely to do average, but on average you will ALWAYS do higher than the Amy.

Points to the Djinn for its weapon.
Arc's can go fuck themselves, because if you can't get the enemy in to a F/S arc you shouldn't be playing dropfleet.

Right, R-squared for people who can't math, R-squared is the likelihood of the next number on the graph falling within the average values given, at least in this case, What we want is a High R-squared and a high average attack, a weapon with a R-squared of 1, will ALWAYS do average damage, a good example of this is the diamonds particle triad, about 99% of the time, it will do a minimum of 3 damage, the lower the R-squared value, the more likely the weapon is to do a non-average amount of damage (Not specifically lower, or higher, just not average) A very good example of this are the UCM's BTL's, with a R-squared of something like 0.30, or Torpedos, with the lowest R-squared in the entire game.

However! The Djinn has its 'closing' weapon, which while puny is still a further increase in its firepower, meaning that over-all the Djinn just edges out the Amy in overall capability, mathematically at least. So, why take the Amy at this point? Probably because people are scared shitless of shaltari and will scream and whine whenever the Shaltari get anything interesting or unusual to play with. but when something of theirs is so crap it never see's the light of day no one bats a eyelid (I'm looking at you Aquamarines!). And Amethysts are the epitome of this. So people make rash, unwise decisions that impact their gameplay when they see a Amy CAW pack blazing down their gullet. Stay cool, think carefully and execute key targets. Its like chess, or risk, A calm head, a good grasp of the mechanics, and a fundamental understanding basic physics and geometry will lend you victory. (Unless you play with Jades or Shaltari Light cruisers, or any non-heavy particle ship, in which case prepare your anus)

I am good at memorizing numbers and tables so to me it's not an issues.

So you can memorise all possible outcomes of 4 Djinns firing? Compared to all the possible outcomes if we change this to 2 firing at one and 2 firing at another target? Impressive if true.

Even if it is only the first one we are looking at a large amount of probabilities.
The Number of CAW attacks vary between 12-32, then we track each single one of them with all possible results Crit/Hit/Miss, then we add to that the 4 Occulus shots and their probabilities, depending whether or not they are within scald range.
At the last step with we factor into the whole thing the enemy PD. I am sure you memorise all those number Lyraeus certainly you do.

Oh I forgot we have to recalculate all this if modifiers apply.

Im not a math person so if you could show me what one of these charts or tables shows for an Amethyst vs a Djinn I'd be curious to see the difference in the percent numbers vs averages.

Usually there are only minor differences some things that appear identical are in fact slightly different and so on, I would put this at a 5-10% inaccuracy. Essentially demanding to use percentages means that you are not actually calculating certain scenarios 99% of the time as it becomes to much work, also if you want to retain accuracy you have to recalculate a lot if the circumstances change a little.

That is why Lyraeus prefers to be dismissive of "on paper" maths and instead bases it on his personal experience. He also likes to forget that the average damage is just used to compare the damage nothing else. Claiming that the Djinn does a lot more damage on average then the Amethyst does not equal claiming it is the better ship. It is a factor in regards to which of them is the better ship but not the only one. If ships are otherwise identical, or the ship with the worse damage has also the worse "soft stats", then the average damage can tell you which is better.

If you like I am willing to provide an example of Average vs. Percentages for 1 Djinns (before PD) and 1 Amethyst which is enough effort. You could of course multiply the results to simulate more ships but then you are again averaging things.

I am good at memorizing numbers and tables so to me it's not an issues.

So you can memorise all possible outcomes of 4 Djinns firing? Compared to all the possible outcomes if we change this to 2 firing at one and 2 firing at another target? Impressive if true.

Even if it is only the first one we are looking at a large amount of probabilities.
The Number of CAW attacks vary between 12-32, then we track each single one of them with all possible results Crit/Hit/Miss, then we add to that the 4 Occulus shots and their probabilities, depending whether or not they are within scald range.
At the last step with we factor into the whole thing the enemy PD. I am sure you memorise all those number Lyraeus certainly you do.

Oh I forgot we have to recalculate all this if modifiers apply.

Im not a math person so if you could show me what one of these charts or tables shows for an Amethyst vs a Djinn I'd be curious to see the difference in the percent numbers vs averages.

Usually there are only minor differences some things that appear identical are in fact slightly different and so on, I would put this at a 5-10% inaccuracy. Essentially demanding to use percentages means that you are not actually calculating certain scenarios 99% of the time as it becomes to much work, also if you want to retain accuracy you have to recalculate a lot if the circumstances change a little.

That is why Lyraeus prefers to be dismissive of "on paper" maths and instead bases it on his personal experience. He also likes to forget that the average damage is just used to compare the damage nothing else. Claiming that the Djinn does a lot more damage on average then the Amethyst does not equal claiming it is the better ship. It is a factor in regards to which of them is the better ship but not the only one. If ships are otherwise identical, or the ship with the worse damage has also the worse "soft stats", then the average damage can tell you which is better.

If you like I am willing to provide an example of Average vs. Percentages for 1 Djinns (before PD) and 1 Amethyst which is enough effort. You could of course multiply the results to simulate more ships but then you are again averaging things.

If I wanted I could probably memorize the bulk of a chart for hits. The most likely and unlikely ones. I mean, sure Lorn. You know exactly what I am capable of or know if I have an eidetic memory or not. I have a string eidetic memory. I see pictures of things in my head which is why I can memorize tables easily.

I am not "dismissive" of units on paper, I just have enough experience that units on paper are not always the same thing as units on the table. Yeesh, you do enjoy attacking me don't you.

*snips the amazing information out becuase it is Sooooo long yet great! *

First off thank you for the lesson, I learned a lot.

Second, I did the four rolls to show that the averages don't play out every game, not to conduct a black swan argument. It was to show that while averages are helpful we can't not base all expectations off of them as it will take dozens of games to get those.

I personally would give the Armor to the Amy in your scoring. The Amy has the option of getting a better armor and can save crits. The Djinn can not. This is a versatility thing. In games you can use shields when needed and drip them to rudely suddenly be out of range.

One of my fun moments was against Nigel. I was able to wipe out his Echoes before the dropped to atmo and stayed shields off while coming around the flank. I did shoot through debit and used that to make the Amy's invulnerable that turn which is why scan is always worth more to me.

Scan is amazing and is the heart of this game. Try the comm stations mission as Scourge and get both comm stations and tell me it is not. 12" Scald is stupid...

Second, I did the four rolls to show that the averages don't play out every game, not to conduct a black swan argument. It was to show that while averages are helpful we can't not base all expectations off of them as it will take dozens of games to get those.

That is exactly what a black swan argument IS, You're saying that in a debate about averages, across the whole of the game as it is played across the entire planet, 4 rolls are a reasonable sample size.
Look at it this way, In 1 game of dropfleet, how meany dice do you roll? 30? 40? no, its more like 300 or 400 dice if you ask me. I'm saying "across every game ever played this is what is most likely to happen" Hence why I included the R-squared values (or a explanation of) Because I can go "Yes, but over the course of 20 rolls, It is 70% likely that they will fall within the predicted R-squared value, therefor they are not capable of discounting the average." Its called a absolute proof one which cannot be refuted. I have gone and math'd out all the R-squared values for most of my ships, so I know that in any given situation my amy will on average give me 3 damage, however there is a 20% (technically higher because you have to add the 2/6 miss chance) chance that it will give me something different. Now I can go mentally "How absolutely critical is it that I get that minimum 2 damage and am I willing to take that risk." The issue with going "in my personal experience" In any dice game is that can simply be countered by "well in my personal experience the contrary happens" Go play D&D for a few years, You'll learn why the law of averages is so important, and why people are always trying to reduce the number of dice rolls needed, because the higher the % chance of NOT the average happening, the less chance you have of getting what you want, therefore to achieve a valid and reasonable discussion of what is, and isn't a viable option, you MUST talk averages.
That felt kinda rambly but I hope it made sense.


While I would agree with you that Scan is a key component of the game, I would not say it is the Heart. Scan is unchanging, you can compensate for it, Spikes, and Spike management is the core of the game, You could have 6" better scan than me, but if you are always on a minor spike it doesn't matter. but again, You're using incidental evidence to try and argue against empirical evidence. The issue here is the same issue I have with a player in my group. He argues always, from personal experience and what "he has seen" He thinks torpedo's need nerfing because he's only seen them twice used in games, and both times the player using them nailed sixes. So that's what incidental evidence doses for you. You can use it to prove or disprove anything in your own mind.
As for the breakdown and scoring, Honestly it was more to point out that before the weapons and groupings, the ships were pretty much evenly matched. albeit with the Djinn being cheaper.

Second, I did the four rolls to show that the averages don't play out every game, not to conduct a black swan argument. It was to show that while averages are helpful we can't not base all expectations off of them as it will take dozens of games to get those.

That is exactly what a black swan argument IS, You're saying that in a debate about averages, across the whole of the game as it is played across the entire planet, 4 rolls are a reasonable sample size.
Look at it this way, In 1 game of dropfleet, how meany dice do you roll? 30? 40? no, its more like 300 or 400 dice if you ask me. I'm saying "across every game ever played this is what is most likely to happen" Hence why I included the R-squared values (or a explanation of) Because I can go "Yes, but over the course of 20 rolls, It is 70% likely that they will fall within the predicted R-squared value, therefor they are not capable of discounting the average." Its called a absolute proof one which cannot be refuted. I have gone and math'd out all the R-squared values for most of my ships, so I know that in any given situation my amy will on average give me 3 damage, however there is a 20% (technically higher because you have to add the 2/6 miss chance) chance that it will give me something different. Now I can go mentally "How absolutely critical is it that I get that minimum 2 damage and am I willing to take that risk." The issue with going "in my personal experience" In any dice game is that can simply be countered by "well in my personal experience the contrary happens" Go play D&D for a few years, You'll learn why the law of averages is so important, and why people are always trying to reduce the number of dice rolls needed, because the higher the % chance of NOT the average happening, the less chance you have of getting what you want, therefore to achieve a valid and reasonable discussion of what is, and isn't a viable option, you MUST talk averages.
That felt kinda rambly but I hope it made sense.


While I would agree with you that Scan is a key component of the game, I would not say it is the Heart. Scan is unchanging, you can compensate for it, Spikes, and Spike management is the core of the game, You could have 6" better scan than me, but if you are always on a minor spike it doesn't matter. but again, You're using incidental evidence to try and argue against empirical evidence. The issue here is the same issue I have with a player in my group. He argues always, from personal experience and what "he has seen" He thinks torpedo's need nerfing because he's only seen them twice used in games, and both times the player using them nailed sixes. So that's what incidental evidence doses for you. You can use it to prove or disprove anything in your own mind.
As for the breakdown and scoring, Honestly it was more to point out that before the weapons and groupings, the ships were pretty much evenly matched. albeit with the Djinn being cheaper.

You at best get 6 attacks per game. The amount of dice does not matter. Even if you weapons free every single turn you get 6 Attacks.

Tell me, how long do your Amy's live? What about the Djinn's you see? Typically Amy's can live an entire game getting 4 attacks while the Djinn I see don't. I was generous on showing a games worth of shots for the Djinn.

I don't see that as a black swan argument. I really just see it as a "expectations are not always what you get". Call it what you will but I am not saying this with only incidental experience.

I am looking at all of this from my time in the Beta to the numerous games I have played. I made the whole faction ratings based of well over 100 games under my belt. It is not just me playing a few games and rating them.

Well I've spent all day rolling a few million dice and recording them in a spreadsheet for fun... well not really. I spent a little bit of time whipping up a macro in excel to roll the dice and gather the data.

Here are a couple of charts, which are going to prove whatever your point of view was before seeing them. Data is tricksey like that.

So each iteration is rolled 100,000 times. All saves and PD are rolled and applied as appropriate and all random number of shots are rolled separate for each of the 100,000 attacks. If the targets save is 3+ or 4+ it applies PD to crits first, if the save was to be 5+ it i set to apply to PD to hits first. This wasnt needed for what I rolled here, but horray for future proofing.
I have made the images a bit big for these forums, so please be prepared to scroll to the right a bit to see what line belongs to what.


This one explains itself.


Lots of ways to interpret this, you could say Djinn and Taipai vary wildly, and they sure do! I emplor you to see the chances of getting those far reach numbers are and you will need to pick a point at which those outside chances are not worth looking at. Perhaps for you the outside 5% for Djinn are to be relevant. In which case the lowest damage from 4x Djinn reasonably would be 7, while the highest would be 17. and the 4x post nerf Amys range would be lowest damage 2, and highest 8.

Dont forget that while the Taipai sits behind the Djinn on this chart, this dosnt take into account the cost of the vessels and with that in mind the Taipai isnt very far behind the Djinn at all.
Its a bit easier for me as I have access to the tables of damage numbers and percentiles, while you get to read a graph.

So its pretty clear we all agree its a matter of trade off when comparing the Djinn and Amys. And the army they fit in has to be taken into account alongside the 12"scan to 6" and the option of shields or not.
Now if we assumed that the Djinn was working as intended and was a perfect fit in the game then we have a starting point.

Now what if we said we were hunting San Frans and we decided that achieving 8 damage was enough and were going to rely upon cripple to finish it off.
4 Djinn have a 92.97% chance to do this.
4 Taipai have a 62.99% chance to do this.
4 Pre Nerf Amy have a 41.77% chance.
4 Post Nerv Amy have a 12.39% chance.

What if you wanted to not rely upon the cripple or any other ship in your fleet and do the 10 damage straight up(Shaltari have an easier time getting assistance from than scourge - a factor)
4 Djinn - 78.46 %
4 Taipai - 35%
4 Pre Nerf Amy - 12.86%
4 Post Nerf Amy - 1.55%

Lets kick it up a notch, a whole 13hull point Orpheus. That we will pretend has no Calypso friends.
4 Djinn - 42.89%
4 Taipai - 8.4%
4 Pre Nerf Amy - 0.64%
4 Post Nerf Amy - 0.01% (They did 13 damage, 13 times out of 100,000 rolls. how cute)


If killing strike carriers was a problem to the player base. (fine, I think its a case of the meta with this issue isnt running corvettes in bulk yet) Then yes the nerf has helped and reduced the number of corvettes the Amys can take down effectively. (Djinn can still do this almost as well on reach - debris fields only go so far when you came forward 10" on your first turn)

Was the problem Amys clearing a flank and then circling around? ok the nerf makes the chance pretty fairly remote that they can clear their flank by themselves, but it does nothing about the issue of being behind a fleet which cant turn around to reach them. To get through the flank the Amys need a fair amount of help, weather it was killing a San Fran, Seattle and a Jakarta, or just a San Fran they struggle to do that by themselves. Perhaps it would have been better to reduce their scan range to 6" or 8" if this was the problem with Amethysts?

Was the problem Amys ability to knock out critical targets in an unstoppable blow? Yes, the nerf certainly fixed this. But I still think that every faction has the ability to have their list built to provide options for a general to counter it. Not always counter able, but options are there. Its certainly possible to make a list were you have no chance to alpha strike a low SR group. I have recently been enjoying an SR3 Djinn group just for such occasions, if the other guy has an SR4 Djinn/Taipai/Echo/Amy then I have first choice and can alpha strike his low SR group and cripple it if things were timed right for which round for when our groups arrived to the battlefield.

At the end of the day the question is, does the reduced (significantly?) offensive power offset the 12" Scan and the option to raise shields?
And for me, I think that the dps nerf goes too far, if it was needed at all. I think it is a shame that the Amethysts do not pack enough punch to down targets by themselves (they are the most expensive CA Frigate after all).


If the Amethyst needed a touch up, I'd look to the 12" Scan first, and drop it to 8" and leave its original damage output alone as there was nothing impressive about its damage output and its pretty dismal now.




Just for the fun I ran one to see the effect 4 Jakarta or 6 UCM Fighters have on Djinn.



I'm pretty sure no opinions will be changed as a result of this discussion.

Never said the math did not support you Kemble bit you are still ignoring ranges. Well under valuing them. The Djinn can never escape retaliation. Not ever. Hell it can even be caught up in the explosion.

That 12" scan more important than you think. Play with some Comm stations and use your Djinn later Kemble. See what happens

Djinn can be caught up in explosions? yup sure can. I'll help the odds out and say its an Avalon going kaboom. Its pretty easy to keep djinn 5" away from your target and maintain more than 2" between the Djinn. There is less than 12.4% chance that a Djinn within 5" will face a cripple result.

No I'm not ignoring the 12" scan, I just value it differently.

You put more value in the scan range and believe that doing 4 to 8 damage in 1 volly from 4 Amethysts is a rocking good ship for 196 points.
I dont.
You think they survive significantly longer than Djinn, I dont, I think they can survive longer but not significant enough to justify the lowered damage output. The Amy would have to fire twice untouched and at full capacity and still not match what the Djinn do in just 1 volley. We shouldnt pretend that Djinn all die after their 1 volley every time either, it depends upon where you put them.

Your standard game size will play a factor here as well. If you are playing at 1250 or lower than there will be fewer ships around to retaliate. My meta plays 1500, so perhaps that helps me see Amethysts only survive a little better than Djinn. Maybe my meta is bigger on Corvettes and other CA Frigates to counter punch Amethysts or Djinn to ensure they dont get to strike multiple times or at least not at full strength. Maybe you play 1500 as well or even higher, I dont know.


I value the ability of Djinn and Taipai to a lessor extent to clean house in 1 round and not face return fire from whatever the target was.
And if you can pull the holy grail and do your first strike as the last activation of a round and get strike again as the first. Then the Djinn can clean up a flank of their own and face no return fire. While an Amy need both of those rounds to clean up a single cruiser target unless they get help. Help that Djinn dont need. I value the fact Djinn dont need help.



Its like you think I've played the game twice. Yes the comms station mission, I've played quite a few times. I've had control of both stations. Your opponent dosnt make that mistake twice and let scourge have the comms stations. Of course we could be talking about an 18" scan Amy here. Oh the glory.


If the Amethyst needed a nerf I'd rather see it with an 8" Scan range and have its original attack dice. Than a 12" Scan and the current experimental attack dice.

Last post and I'm going to try really hard to stay away as we are surely at the point where you just say "Scan Range" and I spend 20mins writing constructive points to show why I believe that the nerf to Amethysts isnt a good solution.

From my point of view, the damage output nerf to Amethysts prevents them from being reasonably used as bomb units and taking out critical targets upon request. Going to 6 Amethysts (300pts) to do less damage than 3 Djinn (129pts) but I digress, 6 Amethysts to give yourself a decent chance at killing a standard cruiser is not viable to me. SR4 Djinn / Taipai packs are too common, as are SR4 Echos and SR5 or 6 Nickar and New Orleans groups. SR6 just isnt reliable for this function.

And without the ability to be used as a bomb unit, it divulges into something that runs around tickling ships with a 360* fire arc and only 12" Range. At that point I'd rather a Topaz if taking frigates, at least it gets to use sigs and spikes and its damage output is only slightly below the nerfed Amethyst.