I have been working on an improved edition of Savage Attacker, as I really like the concept of the feature, yet after your class get's more than one attack, it feels kind of lackluster.
So I came up with an Improved version here on DnD Beyond and I'm looking for your feedback.
For your ease, here is the text:
This warrior has learned to tap into his brutal side to land more devastating blows. When he swings his/her blade, their foe will tremble for the impact which is incoming.
Increase your Strength or Dexterity score by 1, to a maximum of 20.
Once per round when you roll damage for a melee weapon attack, you can re-roll the weapon its damage dice and choose the higher result. If the rerolled damage <= the initial roll, you may add half the maximum of the weapon damage dice to this result and pick that sum.
Once per round on a critical hit on a melee weapon attack, you may reroll the damage dice (but not add half max damage dice).
How do you calculate the half of maximum weapon damage? For d12, 2d6 and 3d4 this is 6. For d10 or 2d5 this is 5 and so on.
So for example you are fighting with a 2d6 weapon and you rolled 8 for damage. You choose to use the Savage Attacker re-roll and you roll a 5. This wille become 11 (=5+6) instead and will give you a +3 bonus damage.
At first this seems a little clunky but once you have actually made the roll, it's pretty simple. In the RAW feature you would be disappointed if your re-roll would end up equal or lower than your initial result, now you get to add half of the maximum of your weapon dice!
I have kept the part that it can only be used one, because discrete choices make games interesting: you feel like you have more of an impact on the flow of the game. I just allowed for the re-roll to be made during a whole turn, to create more dramatic effect if a creature triggers an AoO. I also added in the extra boost of damage, so you have a more reliable damage output on one attack a turn. The critical part is there to avoid feeling bad about making an actual critical with a savage character.
I added in the half ASI because I felt there needed a little boost that is always on, so your feature wouldn't feel useless when you don't have that many encounters and fluffwise it seems to fit. As a Savage character also typically has a more outspoken physical side to him.
I hope I didn't went overboard, because I feel that's the trap of home-brewing a feature: you end up creating your wish-list instead of having something balanced.
I am aiming for top to good tier feature but not as insane OP like Sharpshooter or GWM.
My actual build in which the need for this feature was born. I you want to know, I came onto Savage Attacker because I am currently playing a Half-Orc Barbarian but I wanted a good reason to wield a Great-Axe (GA). In the current rules it feels like picking a Great-axe is a bad choice. It's a dicey choice as your damage output is not as much guaranteed as for the Great-Sword. Being a Half-Orc does make the GA swing like before: having a critical x3 yet just the chance for a critical are too low to justify that choice in my opinion. As a tank I want to be able to dish out a reliable amount of damage that is big enough not to be ignored. I could go straight up for GWM but I want to postpone that for around lvl 10.
I am currently at lvl 5 barbarian and have the standard array, in which I have pushed STR and CON as main stats (both 16) and had DEX sitting at 13. So I must admit that I was looking at many half ASI features, to get that 1 point boost at DEX. I have picked the subclass of Ancestral Guardian as it fits well with my character background and it feels like a good tanky path. After level 6 I intend to dip 3 levels into Fighter to obtain GW fighting style and Champion, to go crit fishing. After that, I want GWM to cash in on criticals and more damage output.
The whole "once per round, if you rolled good damage, do over max damage" part is a little strong. And over all it is worded clumsily.
Thanks for the feedback.
What do you mean by "do over max damage", I have designed it like: have a chance to do over max damage. It's about a 50% chance to add damage compared to your initially rolled damage. I don't know if this is too strong, if you compare it to GWM or SS, which allows you to add +10 for a -5 to hit penalty. With this version of savage attacker you can only get a maximum of +6 damage (on a 1d12 or 2d6 weapon) if you roll exactly the same result in the reroll.
About the wording: what part is not clear? Maybe I should remove the symbols for smaller than or equal to and replace them by words?
The thing is I have been looking for too long at the wording, so I don't have a fresh perspective on the wording as I know what I am trying to convey.
Interesting rework you've come up with. A little bit complicated, but the concept is clear. I've done my own tinker with this sub-par feat and I came up with a similar version. Mind if I share it on this thread?
SAVAGE ATTACKER
You've learned to channel your aggression into your attacks, empowering your strikes with savage might.
Once per turn when you roll damage for a melee weapon attack, you can add a bonus equal to half the number of sides on one of the weapon's damage dice to the attack's damage.
Examples.
Example #1: A 5th-level Barbarian makes two attack rolls on his turn with a Greataxe and hits his target with each attack. On his damage rolls (1d12 each), he rolls a 3 and a 7 for a total of 10 damage on his turn. He can then decide to use his Savage Attacker feat to add a +6 bonus to his first damage roll, increasing the attack's damage from 3 to 9 and his total damage from 10 to 16.
Example #2: A 5th-level Fighter makes two attack rolls on his turn with a Greatsword and hits his target with each attack. On his damage rolls (2d6 each), he rolls a 3 and a 7 for a total of 10 damage on his turn. He can then decide to use his Savage Attacker feat to add a +3 bonus to his first damage roll, increasing the attack's damage from 3 to 6 and his total damage from 10 to 13.
Example #3: A 4th-level Rogue makes one attack roll on his turn with a Rapier and hits his target with the attack. On his damage roll (1d8 + 2d6 Sneak Attack), he rolls a total of 10 damage (2 + 5 + 3). He can then decide to use his Savage Attacker feat to add a +4 bonus to his damage roll, increasing the attack's damage from 10 to 14.
Feat Creation Process.
My first attempt emulated the half-orc's Savage Attacks feature by granting an the warrior an extra weapon damage dice once per turn. However, after doing the math, I concluded this seemed a little overpowered as it essentially gave the warrior a free critical hit every round. (Actually, he'd get two free critical hits if he lands an opportunity attack every round, now that I think about it.) This extra damage would make the warrior overpowered against appropriate CR monster, leading to abnormal short fights.
After a little more fiddling around with the mechanics, I came up with my current version of the feat, which gives the warrior a flat damage bonus rather more damage dice. Mechanically, my current version of the feat deals the same average weapon damage as my previous version, but it reigns back the maximum weapon damage while also bolstering up the minimum weapon damage. Essentially, it gives the warrior a "half critical hit", which I feel is a much more balanced (and unique) ability than just copying over the half-orc's Savage Attacks feature.
Overall, this revised feat feels like a lighter version of the Great Weapon Master feat (e.i. lower damage bonus, but without a penalty), which I think is a balanced, flavorful feat that I wouldn't mind playing around with one of my own characters.
I commented on the actual feat itself what I thought, but as a summary SA isn't nearly as bad as most people think it is...is it worse than just taking an ASI? Yes...but most feats are, even GWM if you actually do the math compared to a 2 point boost to strength. Simply granting a +1 to your STR or DEX score in addition to what SA already does would get it in line with other feats though...your suggestion goes too far imo, and as written let's you reroll all damage dice on a crit...like there aren't many classes that won't want the ability to reroll everything when they crit...Hexblade Warlock's, Paladin's, Fighter's, Barbarian's...like any martial class is going to want that. But it goes too far without the possibly unintentional verbiage used for critical strikes.
@MukPile. You version is very similar because it also uses half the max damage total of the weapon dice to add to damage output once per turn.
It is a lot more easy, yet I feel it allows more to 'break' the natural damage limit of an attack. You can roll max damage and add half damage on top of it. So in effect creating a critical hit, in damage output. Doing 150% damage out of the potential 200% is a very goodcritical roll, yet you still need to roll max damage for that but to arrive at 100% crit, which is 100% max damage dice, you only need to roll on the median value (middle value).
I is also a more reliable way of avoiding low damage rolls as you can once per turn simply add 50% to it. Which makes it at least a decent damage roll.
So in all this becomes a very reliable feat to step-up your damage output per round, by 50% of max damage.
The thing that I dislike is that it favors the Greatsword, which is already favoured, in my opinion. Because of it's more gauss curve distribution, you already have a small chance of rolling low on damage. The disadvantage is that you of course have as few high rolls as well. This feature allows you to have your average rolls turn into a high roll guaranteed. Combine this with Heave Weapon Fighting style (re-rolling 1's and 2's) and you'll be golden with a Greatsword.
This is more a personal flavour issue because I feel in 5e the Greataxe has lost it's appeal or niche. Only the Half-Orc Savage Attacks features allows the greataxe to shine on a barbarian only, becaue they reinforce the critical hit more. Yet you only roll a critical 5% of the time. Which means if you have less frequent combat, say about two instances each session, you might go through 4 sessions or more without seeing any effect of it.
But a very nice way to fix a feature that has a good idea behind it but feels poorly playtested. Especially because I have the feeling that in 5e they are moving in the right direction: rewarding players for making a certain choice. So having a feature that potentially has no impact, is rather sad.
I commented on the actual feat itself what I thought, but as a summary SA isn't nearly as bad as most people think it is...is it worse than just taking an ASI? Yes...but most feats are, even GWM if you actually do the math compared to a 2 point boost to strength. Simply granting a +1 to your STR or DEX score in addition to what SA already does would get it in line with other feats though...your suggestion goes too far imo, and as written let's you reroll all damage dice on a crit...like there aren't many classes that won't want the ability to reroll everything when they crit...Hexblade Warlock's, Paladin's, Fighter's, Barbarian's...like any martial class is going to want that. But it goes too far without the possibly unintentional verbiage used for critical strikes.
Many thanks for waying in on this discussion. You are right, that I haven't stipulated that the re-roll on the critical is only for the weapon damage dice. That's sloppy of me. I'll edit it.
How are combat round structured currently? I don't see how the current version of the rules lets you use Savage Attacker feature as it is outside of your player turn. As I see it a combat round is made up by each character taking it's turn, in order of initiative. Correct me if I'm wrong, because this is a rather big impact if I'm wrong.
GWM analysis and comparing it to ASI Yet on the analysis that the feature would be good enough if it grants you one weapon damage dice re-roll and half an ASI, I don't agree. I have done the analysis comparing it to a normal ASI and in pure damage, the ASI wins because you will do more than one attack and all attacks are influenced by ASI.
The GWM feature isn't something that I would like to analyse without the notion that you can gain Advantage quite easily on the attack rolls, which nullifies the -5 to hit part. Knowing that, the benefit of adding +10 damage is huge. What is the maximum average damage of a melee weapon? 7. Adding 10 damage = 142% more average damage, that equals 1,4 extra attacks. So if you can attack twice a turn, you're in equivallence attacking for 4,84 average attack hits. I can't see how any ASI can come near to this damage potential. Of course, I have used advantage to mitigate the to-hit penalty and you can only inflict damage when you hit, so this is very important not to forget. Yet the potential is there and you still have the added bonus of gaining a bonus attack on a critical hit and on dropping an opponent. So the high roll potential is to do 6,26 times average damage on a turn when you gain a bonus attack (not critical), on critical it becomes 7,26.
I commented on the actual feat itself what I thought, but as a summary SA isn't nearly as bad as most people think it is...is it worse than just taking an ASI? Yes...but most feats are, even GWM if you actually do the math compared to a 2 point boost to strength. Simply granting a +1 to your STR or DEX score in addition to what SA already does would get it in line with other feats though...your suggestion goes too far imo, and as written let's you reroll all damage dice on a crit...like there aren't many classes that won't want the ability to reroll everything when they crit...Hexblade Warlock's, Paladin's, Fighter's, Barbarian's...like any martial class is going to want that. But it goes too far without the possibly unintentional verbiage used for critical strikes.
Many thanks for waying in on this discussion. You are right, that I haven't stipulated that the re-roll on the critical is only for the weapon damage dice. That's sloppy of me. I'll edit it.
How are combat round structured currently? I don't see how the current version of the rules lets you use Savage Attacker feature as it is outside of your player turn. As I see it a combat round is made up by each character taking it's turn, in order of initiative. Correct me if I'm wrong, because this is a rather big impact if I'm wrong.
GWM analysis and comparing it to ASI Yet on the analysis that the feature would be good enough if it grants you one weapon damage dice re-roll and half an ASI, I don't agree. I have done the analysis comparing it to a normal ASI and in pure damage, the ASI wins because you will do more than one attack and all attacks are influenced by ASI.
The GWM feature isn't something that I would like to analyse without the notion that you can gain Advantage quite easily on the attack rolls, which nullifies the -5 to hit part. Knowing that, the benefit of adding +10 damage is huge. What is the maximum average damage of a melee weapon? 7. Adding 10 damage = 142% more average damage, that equals 1,4 extra attacks. So if you can attack twice a turn, you're in equivallence attacking for 4,84 average attack hits. I can't see how any ASI can come near to this damage potential. Of course, I have used advantage to mitigate the to-hit penalty and you can only inflict damage when you hit, so this is very important not to forget. Yet the potential is there and you still have the added bonus of gaining a bonus attack on a critical hit and on dropping an opponent. So the high roll potential is to do 6,26 times average damage on a turn when you gain a bonus attack (not critical), on critical it becomes 7,26.
The rules for savage attacker are as follows
Savage Attacker Once per turn when you roll damage for a melee weapon attack, you can reroll the weapon’s damage dice and use either total.
Now a turn isn't just your turn, but any turn within the initiative round as you stated. As such you can use Savage Attacker on your turn, or on another person's turn as an opportunity attack...it makes the default stronger but situational. If it were only usable on your turn it would state as such though, which the feat does not. This also means things like Sneak Attack are applied to things like Opportunity Attacks (if the conditions are met still) which is an important thing to keep in mind when running the game.
I will find the math on GWM in a bit and reply again.
Now, a less abstracted way to describe bounded accuracy is to state that on average, if the DM is giving you balanced encounters, your player characters are going to be hitting 60 to 65% of the time. Using the above chart that means on average your minimum die roll to hit will vary between being an 8 or a 9. Now, when using GWM you are shifting your percentage to hit down, which inversely makes the value to be rolled increase to either a 13 or 14, which is a 40 to 35% chance to hit on average respectively. With advantage, you have a 88% chance to roll an 8 or better, and an 84% chance to roll a 9 or better. You only have a roughly 64% chance to roll a 13 or better and a ~58% chance to roll a 14 or better.
Using these values, we can calculate the average damage dealt after 100 attacks with and without advantage, for both an ASI increase and for GWM. We will use a Greataxe for the calculation and a +3 for the base mod with the ASI increasing it to +4 to start and remembering that the ASI increases the percentage points as well by lowering the needed roll. With the ASI, after 100 attacks and assuming perfect averages and standard attacks, needing to roll 7 or higher would result in 735 damage being dealt, while needing an 8 or higher would result in 682.5. For GWM, you'd get 780 if needing to roll a 13 or better and 682.5 if needing to roll a 14 or better. Not bad, makes GWM look alright or at least equal to the ASI. Now let's look at rolls with advantage, and on a new paragraph just to make it easier to read.
With needing a 7 to hit with the ASI, you are going to do 955.5 damage, and with needing an 8 to hit 924 damage. With GWM needing a 13 to hit results in 1248, while needing a 14 to hit results in 1131.
So now I am sure you are saying, "Yeah GWM is obviously better by those numbers" but the above represents your DM giving you encounters that range from more on the easier end to deadly in equal distribution which isn't always the case, along with the assumption you will always have advantage in the latter case. But it does allow one to draw the conclusion that the chance of dealing more damage with GWM clearly increases when the enemies are easier to hit. So this would be describing encounters, lets say at forth level, where a character could reasonably have either a +5 or +6 to hit with weapon attacks. So the values I gave would be fighting things where the AC average was either 13 or 14 at that level. Even just 1 more point of AC really changes the damage output of GWM vs the ASI attack though (682.5/585 vs 682.5/630 without advantage, and 1131/994 vs 924/882 with) making the ASI more desirable when you have a DM that gives you less encounters per adventuring day, but makes the enemies encountered tougher. If you know the DM though, and you know they tend to keep the AC's of what you are fighting lower or within the range I described, and you have reliable ways of getting advantage for every attack, then GWM is a great pick. On the other hand if you don't have reliable ways to get advantage it becomes barely better than the ASI in terms of just pure damage output, and if the DM has you fighting things with slightly higher AC while you don't have reliable ways of getting and keeping that advantage, then the ASI in just terms of damage is better.
Finally, the biggest thing to realize is that the ASI increase to strength has more value than raw damage output...it improves your strength saves and checks on top of keeping up with or slightly surpassing GWM depending on the type of DM you have. Because with the numbers provided, for every 100 points that GWM is dealing above the ASI, you are gaining a +1 to average damage to your attacks vs the ASI...significant over time, but at the expense of worse strength saves and ability checks.
In summary, GWM is good or bad depending on how strong the things the DM throws at you, and what you give up in terms of damage is typically equal to what you gain from improved saves and ability checks. If all your character cares about is how hard they can hit things and care less about saves and ability checks, taking GWM will probably net positive results if you are running a barbarian that can have advantage all the time via reckless attack. If running a fighter where advantage is more situational, it may not be worth it to impose worse saves and ability checks for a slight damage bump.
EDIT: The only other thing that may push up GWM above the feat is the ability to make a weapon attack as a bonus action when you crit or kill which i did not factor in.
I think you're forgetting one important point, the story. What does the original feat as it is add to the story? It increases that chance of a martial character doing big damage in a key moment to create a memorable event in the story. Opportunity for great story telling. If your character chews through every encounter without any real risk, then it gets pretty boring.
As for your version, I think it tries to minimize downside. Without a downside it makes the feat an autopick over ASI. If anything it just highlights how GWS needs a bit more downside to bring it in line with the rest of the feats. So therefore is overpowered.
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
Hey DnD enthusiasts!
I have been working on an improved edition of Savage Attacker, as I really like the concept of the feature, yet after your class get's more than one attack, it feels kind of lackluster.
So I came up with an Improved version here on DnD Beyond and I'm looking for your feedback.
For your ease, here is the text:
At first this seems a little clunky but once you have actually made the roll, it's pretty simple. In the RAW feature you would be disappointed if your re-roll would end up equal or lower than your initial result, now you get to add half of the maximum of your weapon dice!
I have kept the part that it can only be used one, because discrete choices make games interesting: you feel like you have more of an impact on the flow of the game. I just allowed for the re-roll to be made during a whole turn, to create more dramatic effect if a creature triggers an AoO. I also added in the extra boost of damage, so you have a more reliable damage output on one attack a turn.
The critical part is there to avoid feeling bad about making an actual critical with a savage character.
I added in the half ASI because I felt there needed a little boost that is always on, so your feature wouldn't feel useless when you don't have that many encounters and fluffwise it seems to fit. As a Savage character also typically has a more outspoken physical side to him.
I hope I didn't went overboard, because I feel that's the trap of home-brewing a feature: you end up creating your wish-list instead of having something balanced.
I am aiming for top to good tier feature but not as insane OP like Sharpshooter or GWM.
My actual build in which the need for this feature was born.
I you want to know, I came onto Savage Attacker because I am currently playing a Half-Orc Barbarian but I wanted a good reason to wield a Great-Axe (GA). In the current rules it feels like picking a Great-axe is a bad choice. It's a dicey choice as your damage output is not as much guaranteed as for the Great-Sword.
Being a Half-Orc does make the GA swing like before: having a critical x3 yet just the chance for a critical are too low to justify that choice in my opinion. As a tank I want to be able to dish out a reliable amount of damage that is big enough not to be ignored. I could go straight up for GWM but I want to postpone that for around lvl 10.
I am currently at lvl 5 barbarian and have the standard array, in which I have pushed STR and CON as main stats (both 16) and had DEX sitting at 13. So I must admit that I was looking at many half ASI features, to get that 1 point boost at DEX.
I have picked the subclass of Ancestral Guardian as it fits well with my character background and it feels like a good tanky path.
After level 6 I intend to dip 3 levels into Fighter to obtain GW fighting style and Champion, to go crit fishing. After that, I want GWM to cash in on criticals and more damage output.
If people are interested, I can share som match about what he actual impact is for the feature on the damage roll.
The whole "once per round, if you rolled good damage, do over max damage" part is a little strong. And over all it is worded clumsily.
Thanks for the feedback.
What do you mean by "do over max damage", I have designed it like: have a chance to do over max damage. It's about a 50% chance to add damage compared to your initially rolled damage.
I don't know if this is too strong, if you compare it to GWM or SS, which allows you to add +10 for a -5 to hit penalty. With this version of savage attacker you can only get a maximum of +6 damage (on a 1d12 or 2d6 weapon) if you roll exactly the same result in the reroll.
About the wording: what part is not clear? Maybe I should remove the symbols for smaller than or equal to and replace them by words?
The thing is I have been looking for too long at the wording, so I don't have a fresh perspective on the wording as I know what I am trying to convey.
Interesting rework you've come up with. A little bit complicated, but the concept is clear. I've done my own tinker with this sub-par feat and I came up with a similar version. Mind if I share it on this thread?
Examples.
Example #1: A 5th-level Barbarian makes two attack rolls on his turn with a Greataxe and hits his target with each attack. On his damage rolls (1d12 each), he rolls a 3 and a 7 for a total of 10 damage on his turn. He can then decide to use his Savage Attacker feat to add a +6 bonus to his first damage roll, increasing the attack's damage from 3 to 9 and his total damage from 10 to 16.
Example #2: A 5th-level Fighter makes two attack rolls on his turn with a Greatsword and hits his target with each attack. On his damage rolls (2d6 each), he rolls a 3 and a 7 for a total of 10 damage on his turn. He can then decide to use his Savage Attacker feat to add a +3 bonus to his first damage roll, increasing the attack's damage from 3 to 6 and his total damage from 10 to 13.
Example #3: A 4th-level Rogue makes one attack roll on his turn with a Rapier and hits his target with the attack. On his damage roll (1d8 + 2d6 Sneak Attack), he rolls a total of 10 damage (2 + 5 + 3). He can then decide to use his Savage Attacker feat to add a +4 bonus to his damage roll, increasing the attack's damage from 10 to 14.
Feat Creation Process.
My first attempt emulated the half-orc's Savage Attacks feature by granting an the warrior an extra weapon damage dice once per turn. However, after doing the math, I concluded this seemed a little overpowered as it essentially gave the warrior a free critical hit every round. (Actually, he'd get two free critical hits if he lands an opportunity attack every round, now that I think about it.) This extra damage would make the warrior overpowered against appropriate CR monster, leading to abnormal short fights.
After a little more fiddling around with the mechanics, I came up with my current version of the feat, which gives the warrior a flat damage bonus rather more damage dice. Mechanically, my current version of the feat deals the same average weapon damage as my previous version, but it reigns back the maximum weapon damage while also bolstering up the minimum weapon damage. Essentially, it gives the warrior a "half critical hit", which I feel is a much more balanced (and unique) ability than just copying over the half-orc's Savage Attacks feature.
Overall, this revised feat feels like a lighter version of the Great Weapon Master feat (e.i. lower damage bonus, but without a penalty), which I think is a balanced, flavorful feat that I wouldn't mind playing around with one of my own characters.
I commented on the actual feat itself what I thought, but as a summary SA isn't nearly as bad as most people think it is...is it worse than just taking an ASI? Yes...but most feats are, even GWM if you actually do the math compared to a 2 point boost to strength. Simply granting a +1 to your STR or DEX score in addition to what SA already does would get it in line with other feats though...your suggestion goes too far imo, and as written let's you reroll all damage dice on a crit...like there aren't many classes that won't want the ability to reroll everything when they crit...Hexblade Warlock's, Paladin's, Fighter's, Barbarian's...like any martial class is going to want that. But it goes too far without the possibly unintentional verbiage used for critical strikes.
@MukPile.
You version is very similar because it also uses half the max damage total of the weapon dice to add to damage output once per turn.
It is a lot more easy, yet I feel it allows more to 'break' the natural damage limit of an attack. You can roll max damage and add half damage on top of it. So in effect creating a critical hit, in damage output. Doing 150% damage out of the potential 200% is a very goodcritical roll, yet you still need to roll max damage for that but to arrive at 100% crit, which is 100% max damage dice, you only need to roll on the median value (middle value).
I is also a more reliable way of avoiding low damage rolls as you can once per turn simply add 50% to it. Which makes it at least a decent damage roll.
So in all this becomes a very reliable feat to step-up your damage output per round, by 50% of max damage.
The thing that I dislike is that it favors the Greatsword, which is already favoured, in my opinion. Because of it's more gauss curve distribution, you already have a small chance of rolling low on damage. The disadvantage is that you of course have as few high rolls as well. This feature allows you to have your average rolls turn into a high roll guaranteed.
Combine this with Heave Weapon Fighting style (re-rolling 1's and 2's) and you'll be golden with a Greatsword.
This is more a personal flavour issue because I feel in 5e the Greataxe has lost it's appeal or niche. Only the Half-Orc Savage Attacks features allows the greataxe to shine on a barbarian only, becaue they reinforce the critical hit more. Yet you only roll a critical 5% of the time. Which means if you have less frequent combat, say about two instances each session, you might go through 4 sessions or more without seeing any effect of it.
But a very nice way to fix a feature that has a good idea behind it but feels poorly playtested. Especially because I have the feeling that in 5e they are moving in the right direction: rewarding players for making a certain choice. So having a feature that potentially has no impact, is rather sad.
Many thanks for waying in on this discussion.
You are right, that I haven't stipulated that the re-roll on the critical is only for the weapon damage dice. That's sloppy of me. I'll edit it.
How are combat round structured currently?
I don't see how the current version of the rules lets you use Savage Attacker feature as it is outside of your player turn. As I see it a combat round is made up by each character taking it's turn, in order of initiative.
Correct me if I'm wrong, because this is a rather big impact if I'm wrong.
GWM analysis and comparing it to ASI
Yet on the analysis that the feature would be good enough if it grants you one weapon damage dice re-roll and half an ASI, I don't agree. I have done the analysis comparing it to a normal ASI and in pure damage, the ASI wins because you will do more than one attack and all attacks are influenced by ASI.
The GWM feature isn't something that I would like to analyse without the notion that you can gain Advantage quite easily on the attack rolls, which nullifies the -5 to hit part. Knowing that, the benefit of adding +10 damage is huge. What is the maximum average damage of a melee weapon? 7. Adding 10 damage = 142% more average damage, that equals 1,4 extra attacks. So if you can attack twice a turn, you're in equivallence attacking for 4,84 average attack hits.
I can't see how any ASI can come near to this damage potential.
Of course, I have used advantage to mitigate the to-hit penalty and you can only inflict damage when you hit, so this is very important not to forget. Yet the potential is there and you still have the added bonus of gaining a bonus attack on a critical hit and on dropping an opponent.
So the high roll potential is to do 6,26 times average damage on a turn when you gain a bonus attack (not critical), on critical it becomes 7,26.
The rules for savage attacker are as follows
Now a turn isn't just your turn, but any turn within the initiative round as you stated. As such you can use Savage Attacker on your turn, or on another person's turn as an opportunity attack...it makes the default stronger but situational. If it were only usable on your turn it would state as such though, which the feat does not. This also means things like Sneak Attack are applied to things like Opportunity Attacks (if the conditions are met still) which is an important thing to keep in mind when running the game.
I will find the math on GWM in a bit and reply again.
Alright, so we are going to be using this table for reference
https://statmodeling.stat.columbia.edu/2014/07/12/dnd-5e-advantage-disadvantage-probability/
Now, a less abstracted way to describe bounded accuracy is to state that on average, if the DM is giving you balanced encounters, your player characters are going to be hitting 60 to 65% of the time. Using the above chart that means on average your minimum die roll to hit will vary between being an 8 or a 9. Now, when using GWM you are shifting your percentage to hit down, which inversely makes the value to be rolled increase to either a 13 or 14, which is a 40 to 35% chance to hit on average respectively. With advantage, you have a 88% chance to roll an 8 or better, and an 84% chance to roll a 9 or better. You only have a roughly 64% chance to roll a 13 or better and a ~58% chance to roll a 14 or better.
Using these values, we can calculate the average damage dealt after 100 attacks with and without advantage, for both an ASI increase and for GWM. We will use a Greataxe for the calculation and a +3 for the base mod with the ASI increasing it to +4 to start and remembering that the ASI increases the percentage points as well by lowering the needed roll. With the ASI, after 100 attacks and assuming perfect averages and standard attacks, needing to roll 7 or higher would result in 735 damage being dealt, while needing an 8 or higher would result in 682.5. For GWM, you'd get 780 if needing to roll a 13 or better and 682.5 if needing to roll a 14 or better. Not bad, makes GWM look alright or at least equal to the ASI. Now let's look at rolls with advantage, and on a new paragraph just to make it easier to read.
With needing a 7 to hit with the ASI, you are going to do 955.5 damage, and with needing an 8 to hit 924 damage. With GWM needing a 13 to hit results in 1248, while needing a 14 to hit results in 1131.
So now I am sure you are saying, "Yeah GWM is obviously better by those numbers" but the above represents your DM giving you encounters that range from more on the easier end to deadly in equal distribution which isn't always the case, along with the assumption you will always have advantage in the latter case. But it does allow one to draw the conclusion that the chance of dealing more damage with GWM clearly increases when the enemies are easier to hit. So this would be describing encounters, lets say at forth level, where a character could reasonably have either a +5 or +6 to hit with weapon attacks. So the values I gave would be fighting things where the AC average was either 13 or 14 at that level. Even just 1 more point of AC really changes the damage output of GWM vs the ASI attack though (682.5/585 vs 682.5/630 without advantage, and 1131/994 vs 924/882 with) making the ASI more desirable when you have a DM that gives you less encounters per adventuring day, but makes the enemies encountered tougher. If you know the DM though, and you know they tend to keep the AC's of what you are fighting lower or within the range I described, and you have reliable ways of getting advantage for every attack, then GWM is a great pick. On the other hand if you don't have reliable ways to get advantage it becomes barely better than the ASI in terms of just pure damage output, and if the DM has you fighting things with slightly higher AC while you don't have reliable ways of getting and keeping that advantage, then the ASI in just terms of damage is better.
Finally, the biggest thing to realize is that the ASI increase to strength has more value than raw damage output...it improves your strength saves and checks on top of keeping up with or slightly surpassing GWM depending on the type of DM you have. Because with the numbers provided, for every 100 points that GWM is dealing above the ASI, you are gaining a +1 to average damage to your attacks vs the ASI...significant over time, but at the expense of worse strength saves and ability checks.
In summary, GWM is good or bad depending on how strong the things the DM throws at you, and what you give up in terms of damage is typically equal to what you gain from improved saves and ability checks. If all your character cares about is how hard they can hit things and care less about saves and ability checks, taking GWM will probably net positive results if you are running a barbarian that can have advantage all the time via reckless attack. If running a fighter where advantage is more situational, it may not be worth it to impose worse saves and ability checks for a slight damage bump.
EDIT: The only other thing that may push up GWM above the feat is the ability to make a weapon attack as a bonus action when you crit or kill which i did not factor in.
I think you're forgetting one important point, the story. What does the original feat as it is add to the story? It increases that chance of a martial character doing big damage in a key moment to create a memorable event in the story. Opportunity for great story telling. If your character chews through every encounter without any real risk, then it gets pretty boring.
As for your version, I think it tries to minimize downside. Without a downside it makes the feat an autopick over ASI. If anything it just highlights how GWS needs a bit more downside to bring it in line with the rest of the feats. So therefore is overpowered.