Yes, that was what I was getting at - I’ve played with the re-creation of this code in openxcom and done some statistical modeling on it. The actual percentage is the chance of rolling a ‘low-deviation’ shot, but then the accuracy again factors into the amount of deviation within that bracket. The end result is that the most intuitive measure I could come up with is that it’s approximately chance to hit a soldier-sized target at 20 tiles, but 0% accuracy still hits ~2-5% of the time at that range. Closer range means better chance to hit for the same accuracy numbers, longer range means lower. Minimum deviation occurs at 110"%" accuracy. Now perhaps it’ll be clearer why giving the numbers aren’t a good choice here?
If that’s how it worked in the original X-Com then whether it’s flawed or not depends on whether how that mechanic worked was explained to the player. I don’t it recall it being explained this way, but it is a long long time ago, and my memory, certainly, is flawed. 
If X-Com’s percentages ‘were’ based on bulls eye shots and that was explained to the player, than I’d personally find that acceptable and understand that the percentage given was a minimum chance to hit. It’s just a question of definition; ‘chance to hit’ can be replaced with ‘minimum chance to hit’. Either way taking an 80% (minimum chance) shot gives a better chance than taking a 20% (minimum chance) shot and I can easily determine that as a player.
As an aside if you’re going to give BS data to the player then better give that data in the player’s favour than against.
Whether you compute the hit percentage in PP using the filled area of a target circle of using a Monte Carlo simulation is secondary to the fact that you’re giving some data to that player that they can work with. Certainly I can gain an equivalent interpretation visually, but I don’t feel that that method gives as accurate information as what the game’s model might estimate for me mathematically. If I’m going to work with estimates I’d rather get a narrow estimate than a wide on. The argument for doing things this way seems to be that because you can’t provide the player with precise information on their chance to hit with a given shot then it’s better to give a rougher estimate have the player blame themselves if they get it wrong rather than give a closer estimate mathematically and risk that they blame the game.
The other downside to only showing visual information is that slows down my game-play if a player needs to continually zoom in and out of shots in order to estimate whether each shot is worth taking or not. Fine if I’m only taking the one shot, but I’m sometimes looking at multiple opponents with multiple soldiers and trying to decide which shot to take first whilst remembering how likely I’d estimated everyone else’s shots at those other opponents. Or alternatively, what I’m doing in practice is looking at the HP indicator to see how much ‘likely’ or ‘possible’ damage. I could be wrong, but I’m working on the basis that ‘likely’ means greater than 50% and ‘possible’ means between 1% and 50%. As to where these ‘likely’ and ‘possible’ indicators are coming from, again I could be wrong, but I’d guess that the game is still doing the Monte Carlo calculations in the background, just now it’s giving me less precise information via that HP indicator than it did previously with its percentages.
2 Likes
SpiteAndMalice is pretty spot on as to what I’m feeling. And don’t get get me wrong, I never asked or pushed for an XCOM style way of calculating percentages.
Since a “to hit” percentage is apparently (for technical and political reasons) not going to work in this game, how about - as SpiteAndMalice was hinting at - a figure that represents something like % of target (center mass) in reticle? Because that is really what I would be trying to visually estimate by zooming in. Of course this figure becomes more or less irrelevant as soon as I zoom in to move where I am aiming, but it would still be very helpful.
And no, there are more percentages than 0% and 100% that matter.
I am so grateful for the lack of shot percentages, it’s true, that’s all you would rely on no matter if it made any sense or not.
In PP YOU; weigh up the shot using your own judgment. Skill, distance, obstacles and such. I find that far more satisfying, and I don’t rage quit if I miss, where as if it had a great big 90% stuck to it, I would. 
Obviously a very political topic. It’s kind of difficult to have a constructive discussion if you assume that everyone will use only the percentage as something to solely rely on and then rage when a high percentage shot misses.
A percentage on center mass in reticle (or similar) would save me a lot of time in considering whether to zoom in for a precision shot or to relocate first. A quality of life setting. My suggestion is therefore to make the above an option to toggle on/off. That way, hopefully nobody needs to feel hurt - unless it hurts already knowing that the option is in the game.
1 Like
This is either mathematically equivalent to the chance-to-hit calculation, assuming that shots are distributed evenly over the circle, if you’re asking for the percent of the reticle filled with the target, or is not very useful if you’re asking for the percent of the target that lies within the reticle, as that would be smaller the closer you are to the target.
Since there is already a simplified calculation of expected damage flashing on the enemy’s health bar, would it be helpful information to have an option to display the expected damage as a percent of the total possible damage of the shot? For example, if you have a Mindfragger egg in the reticle for your shotgun and the health bar is flashing for 20 damage, the number displayed would be 20 / 40 = 50%.
Thanks ohartenstein. Why I would rather call it a quality of life figure/setting is because it would heavily influence whether I find it worth my (IRL!) time to zoom in to look for a good shot. Imagine a target facing me 20 squares away, in half-cover or behind a lamp post; let’s say the % target in reticle indicates 50. With a sniper rifle I’d try to get a good shot for sure - probably in general all the way down to 25% unless what I’m going for is obviously obscured- but with an assault rifle I may not even bother if it’s lower than 75%. Of course situational but still, quality of life more than anything. Not everyone is super hardcore with all the time in the world on their hands It’s not dumbing down the game, it’s speeding up decision-making.
Expected damage percentage… I don’t know. Trying to wrap my mind around it. As far as I understand, what defines damage is the number of hits and the armor of the location (adjusted for weapon damage, damage drop-off over distance, armor piercing and whatever). A percentage may be more pedagogical, I don’t know, but it would not really help out w.r.t. positioning/% target in reticle since I don’t know which of the above factors has what impact on the percentage. If that makes sense? Again, not everyone playing the game will want or be able to memorize the damage of every weapon to assess the damage/armor relation.
In part, we’re trying to make a game that is a little more simulation.
Yes, players are used to having all of the information presented to them so they can make a decision. However, a soldier on the battlefield doesn’t see a floating hit % or health bars above a target when he aims at them. He makes a decision based on his training, experience and instincts.
I’ve seen a lot of people who play Firaxis XCOM, carefully plan out the order of their attacks each turn to ensure they do the minimum damage required to guarantee a kill on each enemy. We have plans in Phoenix Point to actually hide the health bars of a new enemy (the reason being that as you haven’t encountered/killed them before, you don’t know how much health they have). Their health will eventually be revealed through killing enough of them, research, or some form of scanner. No doubt there will be some people who are unhappy with having enemies where they can’t see the health bars so won’t know if their shot is going to be a guaranteed kill or not… and that’s the point. It isn’t really a quality of life issue. It’s a “that’s just part of the game” issue.
3 Likes
This is exactly the issue; chance-to-hit might be easier to understand as a dice roll but is hard to calculate with the bullet mechanics the game is using. Expectation value of a hit in the distribution of shots is an easier calculation and represents what’s actually going on in the game engine, but is harder to intuitively understand. Perhaps instead of expectation value of damage, expected number of hits would be better? This is more or less equivalent to chance-to-hit for single-shot weapons, and gives a better idea of the amount of damage you could do for burst weapons than a chance-to-hit. This would also respect the fact that HP bars won’t be immediately accessible, as UV said.
If the calculation is fast enough, as it seems to be updating quickly when free aiming, perhaps it could also be displayed automatically for the center-of-mass aiming?
The issue with that model is that enemies have different body parts with different armour values. 5 hits from an Assault Rifle against a body part with 1 armour will do 10 damage, but if they hit something with 3 armour will do 0.
1 Like
Right, but that’s part of what we as players may or may not know if we’ve researched the enemy before and know whether it’s armored or not - you have the information that you can expect to hit a few times, but it’s up to your judgement whether or not it’s worth firing based on what you know of the target.
1 Like
I 100% understand and agree with letting game design be the deciding factor!
“He makes a decision based on his training, experience and instincts.” - Sure But the only real difference between displaying a percentage like that and not [EDIT: aside from immersion], is how much time you want the person playing the game to spend on decision-making. If the figure is not presented, I need to zoom in. If I have a lot of time, I will do that every time. If I don’t, or if I’m losing patience, I may go for the random shot. (Random not being entirely random of course, even a bit of game experience will significantly improve understanding of hit percentages.)
I think the idea of hiding health bars is really cool! Health bars would be a start, then we start “unlocking” more stats like location-specific hit points and location-specific armor values.
But what doesn’t change in all of that is how many % of the enemy is in the reticle…
I’m kind of a fan of letting the player customize the experience. You don’t want to scare away new players by making it too confusing or punishing to make decisions based on what they see as too little information. But you also want to cater to the ultra-hardcore players who may not want ANY information more than what is on the screen - no health bars, armor, ammo, whatever. And all those in between, who may want to tweak the experience to match their skill level and play style. You can/should of course have general difficulty levels, but if possible, why not explore the route of putting customization in the hands of the player? That kind of customization will be a BIG USP for this game and may drastically increase replayability for some players.
EDIT: If I have 15 minutes to play a mission, I’ll turn on the % in reticle thing. If I have more time, I may turn it off. Everyone is happy - except perhaps the poor developer who has to implement it 
1 Like
THIS IS EPIC! I want it.
I can confirm, in every single ‘shooter tactic’ my ultimate goal per turn is to kill an enemy unit. Because in 90% of games, there is no difference in damage between 1 and 99% healthy unit. The grenadier with critical hit and bleed on him is as deadly as his fresh companion. That’s why a kill shot is an ultimate goal.
in PP sometimes it’s better to eliminate a weapon/limb of e.g. grenadier to make him easier to swallow.
1 Like
XCOM 2 suffers from this in a pretty big way, mostly because of the Pod mechanic and turn timers. You trigger a pod, and your ultimate goal is to alpha-strike the entire pod down in a single turn so that it doesn’t get a chance to attack. Then you rinse and repeat. I’ve seen people taking 30 minutes on a single turn, weighing up every single possible action combination/order to see which would score the most turn-1 kills.
1 Like
An instant comparison that came to me for PP, and one which I REALLY like, is the combat system in Battletech. Pinpoint attacks can be extremely effective and really reward good positioning. In PP: if you focus fire and disable a leg, the pincer crab is practically out of the game. Destroy the machine gun and the shooter crab is (mostly?) harmless. It adds real tactical depth.
1 Like
But this isn’t XCOM and you don’t have pods, so why bother?
+1
Options, in general, are the way to go in single player games.
Different players like to play games in different ways. I personally would prefer to have a to hit % indicator available, If it was there I’d turn that on. At the same time, I’d turn the predicted damage indicator off because I’d rather work that out for myself. (I’m actually wondering now whether I can backwards calculate the to hit % based on the likely damage indicator and knowing the bullet damage amounts). For that matter I’d also love to be able to turn off the indicator that shows me which tiles give line of sight on various enemies.
My preference overall is for a TBS to be about the tactical positioning and maximising my percentage chances to win in a given situation. As it stands the tactical positioning is being hand fed to me, I just go where the game tells me to in order to make sure that everyone can make a shot. On the flip side, I was then just taking pot shots, I had no idea at all whether those shots were going to hit or not, the best aid I had was the predicted damage indicator because after zooming in on enemies for the first few turns I just couldn’t be bothered with what I considered the fafiness of it.
Granted this was in the early hours of the morning and I was half asleep, but that’s kinda the point, sometimes as a gamer you are switched on and you will invest time to look at non perfect information to make a judgement call, but not all gamers will do that, and those that do won’t necessarily do it all of the time.
But, however a player prefers to play, and however that varies, options serve everyone’s needs.
I remember when the first modern XCom game came out. At first I thought it was pretty good, I was just stoked to see a new Xcom game if I’m honest. But when I saw the second wave options such as ‘Not Created Equally’ and ‘Damage Roulette’, that’s when I became interested enough to make a purchase. I really love the Training Roulette of Enemy Within, which I think goes against the grain as it seems to be that most players leave that option turned off.
You say that like it’s a bad thing…
Yes, this is what I’ve been trying to get across - the to-hit chance is most easily approximated and communicated as expected damage or expected number of hits when we have this projectile simulation - not that we shouldn’t have the option for some kind of indicator for it. I agree with you on wanting some kind of measurement or indicator that appears either when cycling through targets but not free aiming or on the enemy icons when selecting the fire option.
I don’t recall if it’s in the current backer build or not, but didn’t the crabman gunners get a return fire percentage above their icons when the player was aiming at some point? What was this calculation, and is it similar enough to the kind of indicator we want to see that it could be reused?