Jayvee94 Posted January 15, 2016 Share Posted January 15, 2016 (edited) Hi, guys I'm back! I'm sorry I wasn't able to post polls which Riku actually started. But I'm not gonna be able to do that in a while, things are getting busy in college. Anyway to our new topic: If you've been reading the main site, you know that a unit's rating is calculated in game with this formula [spoiler=Formula] Rating = Str + Mag + Skl + Spd + Lck + Def + Res . However, as you all know, not all stats are created equal. Even units with the same in-game rating may have a lopsided outcome due to that one stat. Now, let's expand the formula [spoiler= formula] Rating = 0*Max HP + 1*Str + 1*Mag + 1*Skl + 1*Spd + 1*Lck + 1*Def + 1*Res . As you may see, the coefficients (the numbers that multiply the stat) are 0, 1, 1, 1, 1, 1, 1, 1 respectively. And now, what if we change the coefficients and call it the "TRUE" rating. For example, I will make the coefficients like this: .5, 1.2, 1.0, 0.8, 1.2, 0.8, 1.0, 1.0. It would translate to [spoiler= formula] "True" Rating = 0.5*Max HP + 1.2*Str + 1.0*Mag + 0.8*Skl + 1.2*Spd + 0.8*Lck + 1*Def + 1*Res Let's compare: Let's use Ryoma and Xander as our examples With the in game rating, Ryoma = 113 Xander = 114 Xander won by a small margin there With the Example "TRUE" rating Ryoma = 132.2 Xander = 133 Well, Xander still won by a smaller margin. How about you guys, WHAT WOULD THE COEFFICIENTS YOU WOULD USE TO MAKE YOUR "TRUE" RATING FORMULA PS. Your formula is True from a certain point of view. Edited January 15, 2016 by Jayvee94 Quote Link to comment Share on other sites More sharing options...
CaitlynMellark Posted January 15, 2016 Share Posted January 15, 2016 I'm sorry, but I didn't read this I only came here to screech Because algebra Again, sorry, I just needed to screech algebra Quote Link to comment Share on other sites More sharing options...
Tooru Posted January 15, 2016 Share Posted January 15, 2016 (edited) Def, Res as Asset/Flaw: +10%/-10% HP, Str, Spd as Asset/Flaw: +15%/-10% Mag as Asset/Flaw: +20%/-15% Skl, Lck as Asset/Flaw: 25%/-20% These show how producers rate the coefficients: Def, Res > HP, Str, Spd > Mag > Skl, Lck. But I have some different tiers. Physical + Melee: Str > Spd > Def > HP, Skl > Res, Lck Magical + Melee: Mag > Spd > Str, Skl > Def, HP > Res > Lck Physical + Ranged: Spd > Str > HP, Def, Skl > Res > Lck Magiclal + Ranged: Spd > Mag > Skl > HP > Res, Def > Lck Spd goes the last when there is Defensive Formation. Edited January 15, 2016 by Tooru Quote Link to comment Share on other sites More sharing options...
CocoaGalaxy Posted January 15, 2016 Share Posted January 15, 2016 Eh........it's really hard to tell... Here are mine, which are obviously lazily done Rating = 0.25 HP + STR/MAG (whichever is higher, ignis be damned) + 0.65 SKL + 1.5 SPD + 0.25 LUK + 0.75 DEF + 0.75 RES Obviously it hardly matters......plus it really depends, I dunno. Quote Link to comment Share on other sites More sharing options...
gayserbeam Posted January 15, 2016 Share Posted January 15, 2016 The flaw in multiplying each stat by a constant is that certain stats are more valuable when: 1) other stats are lower / higher 2) the unit has certain skills 3) the unit's class 4) thresholds that need to be met For example: - if a unit has 45 DEF and 45 RES, than each individual point of HP means less to them than it would to a unit with 20 HP and 15 RES. - if a unit is an Axe wielder with Pavise, each individual point of SKL means more to them than it would to a Dagger wielder with no proc skills. - in late game chapters or PVP, when some units are nearing caps, there can be a significant difference between 29 SPD and 31 SPD. However, in the same environment, a unit with 12 SPD and a unit with 8 SPD will both be far too slow anyway, so that stat difference doesn't actually mean the units will perform differently. So yeah, I hate ratings. But that's only because they're impossible to calculate. Quote Link to comment Share on other sites More sharing options...
Parrhesia Posted January 15, 2016 Share Posted January 15, 2016 Unit Rating = its Strength Quote Link to comment Share on other sites More sharing options...
Jayvee94 Posted January 15, 2016 Author Share Posted January 15, 2016 Eh........it's really hard to tell... Here are mine, which are obviously lazily done Rating = 0.25 HP + STR/MAG (whichever is higher, ignis be damned) + 0.65 SKL + 1.5 SPD + 0.25 LUK + 0.75 DEF + 0.75 RES Obviously it hardly matters......plus it really depends, I dunno. If we try your formula and use it on level 4 Sword Master Ryoma and level 4 Paladin XanderRyoma = 103.75 Xander = 97.2 Ryoma wins! Quote Link to comment Share on other sites More sharing options...
Jayvee94 Posted January 15, 2016 Author Share Posted January 15, 2016 We can use the "TRUE" rating formula in determining who is the best in each class. Quote Link to comment Share on other sites More sharing options...
Dunal Posted January 15, 2016 Share Posted January 15, 2016 (edited) It's hard to do because of stats like Speed. Speed on a per-point basis is not that noteworthy if the unit is past the point of doubling, or if that 1 point of speed won't let them double. The main way to rate speed from base stats would be to judge their stat based on who they cannot and can double upon their joining time. Any avoid they get from it doesn't matter as much. Or for another example, the higher DEF/RES a unit has, the higher value their HP will be and vice-versa. A unit with 3 defense and 15 HP will literally get zero gain from extra 1 HP or evel 5 extra HP depending on enemy attack. But someone with 15 DEF? HP going from 15 to 20 can be quite the difference. STR/MAG mean very little if you are 2 or 3HKOing important enemies and one additional STR/MAG would do nothing to change that. OR, if this unit cannot double, then STR/MAG will only increase damage by 1. But with doubling, they gain 2. And the fact that weapon rank affects the importance of having STR/MAG as well. So... as an overall example, you have a unit with 35 speed. You have another unit with 25 speed. Both can double everything else, but the unit with 25 speed also have 2 in every other stat (or also 4 in LCK/HP). In someone's standard grading the guy with 35 speed wins out but aside from the 20 avoid bonus, the other unit is very likely the better unit overall. Edited January 15, 2016 by DLuna Quote Link to comment Share on other sites More sharing options...
Tooru Posted January 15, 2016 Share Posted January 15, 2016 It's hard to do because of stats like Speed. Speed on a per-point basis is not that noteworthy if the unit is past the point of doubling, or if that 1 point of speed won't let them double. The main way to rate speed from base stats would be to judge their stat based on who they cannot and can double upon their joining time. Any avoid they get from it doesn't matter as much. Or for another example, the higher DEF/RES a unit has, the higher value their HP will be and vice-versa. A unit with 3 defense and 15 HP will literally get zero gain from extra 1 HP or evel 5 extra HP depending on enemy attack. But someone with 15 DEF? HP going from 15 to 20 can be quite the difference. STR/MAG mean very little if you are 2 or 3HKOing important enemies and one additional STR/MAG would do nothing to change that. OR, if this unit cannot double, then STR/MAG will only increase damage by 1. But with doubling, they gain 2. And the fact that weapon rank affects the importance of having STR/MAG as well. So... as an overall example, you have a unit with 35 speed. You have another unit with 25 speed. Both can double everything else, but the unit with 25 speed also have 2 in every other stat (or also 4 in LCK/HP). In someone's standard grading the guy with 35 speed wins out but aside from the 20 avoid bonus, the other unit is very likely the better unit overall. It is impossible to have a 35 Spd unit when 25 Spd can double everything else. Quote Link to comment Share on other sites More sharing options...
Dunal Posted January 15, 2016 Share Posted January 15, 2016 (edited) It is impossible to have a 35 Spd unit when 25 Spd can double everything else. I don't think it really matters that your own units can double eachother.. That's nitpicking my overall point, no? Edited January 15, 2016 by DLuna Quote Link to comment Share on other sites More sharing options...
Tooru Posted January 15, 2016 Share Posted January 15, 2016 I don't think it really matters that your own units can double eachother.. That's nitpicking my overall point, no? It is not about "your own units can double each other". 25 Spd can double every enemy -> The fastest enemy has 20 or less Spd -> Mid game -> Our own units won't have 35 Spd. Quote Link to comment Share on other sites More sharing options...
Skarthe Posted January 15, 2016 Share Posted January 15, 2016 It is not about "your own units can double each other". 25 Spd can double every enemy -> The fastest enemy has 20 or less Spd -> Mid game -> Our own units won't have 35 Spd. This will usually be true in practical gameplay, but is also a silly and overly pedantic argument to try to make, since it is not only possible, but its possibility or impossibility in practical gameplay is wholly irrelevant to the argument at hand.The overall point is that Speed is extremely valuable up to the point that you can double every enemy, and then not very valuable past that point. Quote Link to comment Share on other sites More sharing options...
Tooru Posted January 15, 2016 Share Posted January 15, 2016 This will usually be true in practical gameplay, but is also a silly and overly pedantic argument to try to make, since it is not only possible, but its possibility or impossibility in practical gameplay is wholly irrelevant to the argument at hand. The overall point is that Speed is extremely valuable up to the point that you can double every enemy, and then not very valuable past that point. It makes no sense to say "the rate is useless because it cannot be used in impossible situations". Stats, especially Nohr stats, are well balanced in Fates. Whenever a unit can double every enemy, there won't be another unit with +10 Spd than him. The difference in an example shouldn't be as big as "25 Spd" Vs "35 Spd", but something reasonable like "25 Spd + 25 Str" Vs "28 Spd + 22 Str". Quote Link to comment Share on other sites More sharing options...
Skarthe Posted January 15, 2016 Share Posted January 15, 2016 (edited) It makes no sense to say "the rate is useless because it cannot be used in impossible situations". Stats, especially Nohr stats, are well balanced in Fates. Whenever a unit can double every enemy, there won't be another unit with +10 Spd than him. The difference in an example shouldn't be as big as "25 Spd" Vs "35 Spd", but something reasonable like "25 Spd + 25 Str" Vs "28 Spd + 22 Str". As I already said, the gap in stats is unlikely, not impossible - it can be caused by, for instance, extreme RNG one way or the other, stat boosters, or XP grinding (even Conquest can get XP from the appropriate DLC) - and that's still irrelevant. All you're doing is nitpicking the example, not actually countering the point he's making, and the means you're using to nitpick is already an incorrect assumption.And just for fun, Awakening uses the same rating system, and this sort of stat disparity is entirely possible there, especially with a water-tricked Robin. EDIT: Back to the point at hand... What does stat priority look like, anyway? For a combat unit, I'd imagine it'd be something like: Str/Mag to deal nonzero damage > Skl to be able to hit (this one's weird) > Str/Mag to 2HKO > HP+Def/Res to survive a hit > Spd to not get doubled > Str/Mag to OHKO > Spd to double > Def/Res to take 1 damage > further HP > further Skl > further Spd > further Str/Mag > further Def/Res. Skl is weird since I'm not sure how to weight hitting semi-reliably or 100%, Luck is weird since I'm not sure how to weight not getting crit. Edited January 15, 2016 by Skarthe Quote Link to comment Share on other sites More sharing options...
Tooru Posted January 15, 2016 Share Posted January 15, 2016 As I already said, the gap in stats is unlikely, not impossible - it can be caused by, for instance, extreme RNG one way or the other, stat boosters, or XP grinding (even Conquest can get XP from the appropriate DLC) - and that's still irrelevant. All you're doing is nitpicking the example, not actually countering the point he's making, and the means you're using to nitpick is already an incorrect assumption. And just for fun, Awakening uses the same rating system, and this sort of stat disparity is entirely possible there, especially with a water-tricked Robin. You need a lot of stat boosters. Otherwise it is impossible in Fates. Quote Link to comment Share on other sites More sharing options...
GoXDS Posted January 15, 2016 Share Posted January 15, 2016 (edited) You need a lot of stat boosters. Otherwise it is impossible in Fates. for the last time, that's not even the point. it was just an explanation using slightly exaggerated examples. let's take it even more extreme. Unit A has 100 Spd and Unit B has 20 Spd. Unit A and Unit B both double all enemies. who benefits more from 1 Spd? Unit B, yes. now, the whole point of this whole example. how much more does Unit B benefit? barely. or to bring it back to the ratings topic, if Unit A also has 5 in all other stats while Unit B has 10 in all other stats, Unit A's Rating would be higher but you'd probably want Unit B because all that extra Spd Unit A has means absolutely nothing Edited January 15, 2016 by GoXDS Quote Link to comment Share on other sites More sharing options...
Tooru Posted January 15, 2016 Share Posted January 15, 2016 (edited) for the last time, that's not even the point. it was just an explanation using slightly exaggerated examples. let's take it even more extreme. Unit A has 100 Spd and Unit B has 20 Spd. Unit A and Unit B both double all enemies. who benefits more from 1 Spd? Unit B, yes. now, the whole point of this whole example. how much more does Unit B benefit? barely Why must the formula fit the case of "100 Spd vs 20 Spd" + "20 Spd double all enemies"? We can make different formulas for different cases respectively, but the ones for your examples are useless so it is not necessary to waste time for them. Edited January 15, 2016 by Tooru Quote Link to comment Share on other sites More sharing options...
GoXDS Posted January 15, 2016 Share Posted January 15, 2016 (edited) Why must the formula fit the case of "100 Spd vs 20 Spd" + "20 Spd double all enemies"? We can make different formulas for different cases respectively, but the ones for your examples are useless so it is not necessary to waste time for them. the whole pt was that there cannot be a "one formula fits all"... you're way too hung up on the examples. focus on the argument. because there's no one formula fits all, the whole exercise of making a formula means nothing the only time one should ever nitpick the example is if there's a flaw in it such that it doesn't actually prove the argument. otherwise, stick to the argument Edited January 15, 2016 by GoXDS Quote Link to comment Share on other sites More sharing options...
Tooru Posted January 15, 2016 Share Posted January 15, 2016 (edited) the whole pt was that there cannot be a "one formula fits all"... you're way too hung up on the examples. focus on the argument. because there's no one formula fits all, the whole exercise of making a formula means nothing the only time one should ever nitpick the example is if there's a flaw in it such that it doesn't actually prove the argument. otherwise, stick to the argument We don't need to fit all. We just need to fit possible cases in this game, or at lest cases where we need to compare units. A formula may tell us that "a double-everyone unit with Spd 25" is worst than "a unit with +10 Spd but -2 every other stats". It doesn't matter because the latter doesn't exist. "A unit with +2 Spd but -1 every other stats" may exist, and it is more meaningful for comparing. If our formula(s) can give a result which is agreed by most of us, it is useful. There isn't so-called "too much Spd for doubling everyone" in Fates. We may have "sightly more Spd for doubling everyone", which can be transferred to Str by using Steel or something. So even a simple formula may work well in this game. Edited January 15, 2016 by Tooru Quote Link to comment Share on other sites More sharing options...
GoXDS Posted January 15, 2016 Share Posted January 15, 2016 (edited) We don't need to fit all. We just need to fit possible cases in this game, or at lest cases where we need to compare units. A formula may tell us that "a double-everyone unit with Spd 25" is worst than "a unit with +10 Spd but -2 every other stats". It doesn't matter because the latter doesn't exist. "A unit with +2 Spd but -1 every other stats" may exist, and it is more meaningful for comparing. If our formula(s) can give a result which is agreed by most of us, it is useful. Holy lancer KamuiA) 33 Spd because not +Spd and no Yato B) 43 because statues and rally and tonic And +Spd As mentioned already it can exist Formulas are not useful if you can't apply it to a general case. In the case of Fire Emblem, you're likely going to need a new formula for every single comparison and every single situation. That's impractical and imo worthless Edited January 15, 2016 by GoXDS Quote Link to comment Share on other sites More sharing options...
Tooru Posted January 15, 2016 Share Posted January 15, 2016 (edited) Holy lancer Kamui A) 33 Spd because not +Spd and no Yato B) 43 because statues and rally and tonic And +Spd As mentioned already it can exist Formulas are not useful if you can't apply it to a general case. In the case of Fire Emblem, you're likely going to need a new formula for every single comparison and every single situation. That's impractical and imo worthless 33 Spd cannot double everyone so it doesn't exist. A) can also receive rally or tonic so it is not about comparing and choosing units. Edited January 15, 2016 by Tooru Quote Link to comment Share on other sites More sharing options...
GoXDS Posted January 15, 2016 Share Posted January 15, 2016 33 Spd cannot double everyone so it doesn't exist. A) can also receive rally or tonic so it is not about comparing and choosing units. Kanna with +11 Spd mods vs +0 Spd mods. Dlc grinding in early chapter. One or both are from logbook recruit. *sigh* gonna stop caring about your pointless nitpicking if you're not gonna focus on the main debate. Quote Link to comment Share on other sites More sharing options...
Tooru Posted January 15, 2016 Share Posted January 15, 2016 (edited) Kanna with +11 Spd mods vs +0 Spd mods. Dlc grinding in early chapter. One or both are from logbook recruit. *sigh* gonna stop caring about your pointless nitpicking if you're not gonna focus on the main debate. Why not just get a lv99 unit from other My Castle before beating chapter 7 so you don't need to use a lot of stat boosts? Edited January 15, 2016 by Tooru Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.