Dropping Weapons For Bounty Control

[Prog]

On Bounty-Control

1 Preface

This rather short guide focuses’ on the control of anyones own bounty, more precisely I try to find an answer to the question when should I and how can I get my bounty lower when I’m about to die. I’ll present a, in my opinion quite smart, way to do this, which isn’t easy to be used well and which only a few players have in their arsenal of moves. This should be seen as my first attempt to bring a part of my playing style into theory and might have mistakes in maths or general thoughts, so keep your reading critical.


2 Basics

The basic bounty formula for tanks is:

(1) (125 + ( Tankcosts / 65)) * ( ( Kills + 80 ) / ( Deaths +80 ) )


For example a Demolisher with 0-0 stats and Multi Bow has a tankcost of 3000, 0 kills and 0 deaths:

(2) (125+ (3000/65))*((0+80)/(0+80)) = ~171.15

It’s quite evident that the only variable we are able to control or at least manipulate in our favor is the tankcost. We can manipulate kills and deaths as well to keep the bounty lower, but the effect is rather negative, because we either get less money (less kills) or we give our opponent more money (more deaths). Both are not appreciated, so we have to focus on the tankcost. The basic rule we learn from the bounty formula is not “the more items we have, the higher our bounty is”, it’s rather “the more items we have in our inventory when dying, the higher our bounty is”, so we can use this fact to lower our bounty by dropping weapons right before we die. This is the main thesis analyzed in this article, but we have to ask ourselves in which situations it is the correct move to drop weapons (I’ll use weapons in the text synonymous with items).

Let’s see the effect of dropping the Multi Bow in our mentioned example: A Demolisher with 0-0 stats has a Tankcost of 1900, 0 Kills and 0 Deaths:

(3) (125+ (1900/65))*((0+80)/(0+80)) = ~154.32

If we drop our multi bow we deny our enemy 171.15-154.32 = 16.83 (17) gold. This seems to be quite low, a bit higher than 1 marine-creep, but we want to play well, so we need to use small edges. Lets check out a mid game scenario: A Heavy Tank with 8-1 stats and Fire Arrows. Tankcost is 9000, 8 kills and 1 death.

(4) (125+ (9000/65))*((8+80)/(1+80)) = ~286.22

With a drop of the Fire Arrows:

(5) (125+ (5500/65))*((8+80)/(1+80)) = ~227.73

The difference is now 286.22-227.73 = 58.49 (58) gold. As you can see our expected saved value gets higher with the amount of kills and the gold cost of weapons we drop. 58 gold is quite something, as if you die 5 times saving 58 gold each time your opponent gets gold close to 4 times killing you.

3 When to drop

As I showed the weapon drop can save you gold (deny your opponent gold), but there are several aspects to be known concerning the “when to drop”. There are several rules which we have to take care of:

*Of course you shouldn’t drop weapons when the chance you die is zero or quite low, so you have to guess the chance you’ll die.

*As you could see (2) and (3) we saved in this situation approximately one creep in gold. So if we could kill 1 creep instead of dropping the weapon in this situation we got a small positive expected value since our opponent gets +17 gold, while we get for example +25 from the creep we kill, so we got +8 total. If we are able to kill more creeps the value increases with the higher amount. If we can kill the opponent tank its even better. If the weapon we drop is more expensive the amount of creeps we need to kill increases (see (4) and (5)). We have to guess our chances to kill creeps/tanks correctly to be able to adept our behavior to the situation correctly. The only way learning that is by playing and analyzing replays/situations in replays to get kind of feeling for the hp, positions…

*After dropping the weapon we have to get it again, so the weapon has to be at a position we can safely and fast drive to, like close to a control point, or in mid near our towers. If the weapon is far away the opponent can stop us from getting it back and it will need more time (and as the value saved might not be too big it could be too much time to make our drop have positive value: for example we saved +58 gold by dropping, but while we drive to our weapon we could have killed 15 creeps which are worth more. Of course this has to be seen different if another player from your team can kill the 15 creeps (and there are exactly 15, not more) while you get your weapon.

Summary of points which show you when you should drop your weapon:

- You’ll be dead quite certainly
- No or only few creeps around you
- No or very small chance of killing your opponent
- You can reach your weapon easily and fast

At the end I want to show you the thoughts on one situation which occurs quite often:



Let’s check our “rules”:
- You’ll be dead quite certainly: True, the demolisher is level 2 which means level 1 artillery shot, which will kill us.
- No or only few creeps around you: True, there are only allied creeps and enemy tanks.
- No chance of killing your opponent: True
- You can reach your weapon easily: True if we drop our weapon on the way to our control point (approximately where I made the red circle, could be a bit more out of the direct way to the cp to increase or chance of dodging the artillery shot, but it should be not too far away from the control point.

Let’s see how much gold we can save here:
(6) (125+ (3000/65))*((0+80)/(0+80)) = ~171.15
(7) (125+ (800/65))*((0+80)/(0+80)) = ~137.30
= 171.15 – 137.3 = 33.85 (34)



Similar situations occur against demolishers quite often, where the weapon dropping can be used more frequently.



I hope you can learn just a bit from this article. Please mention any critics, especially mistakes in thought or maths. I’d also appreciate you to post screenshots of situations in which you are not sure if you should drop the weapon or not, to analyze it together.

[RedNova]

I'd say I learned a few things in this post but not those intended. Point in case:
- This means that items not in your inventory do not count. Thus, when you are left with no slots availiable (and got cash to spare on extra weapons and items), you can customize your tank with no added cost when you die. For example, you play exploder, but opposing team is in siege. You drop explosives and go for a trader while they are stranded in their base. If they manage to break out, you revert exploder and retake your explosive. So in that situation, the fact that you bought the explosives will not impact your bounty when you die as a trader.

- Some objects do not increase bounty when used. Placed factories, teleport breakers, spent consumables, creep upgrades and others do not count. However, does that include batteries (additional mana), since they are permanently "with you", yet outside the slots?

Personnally, I think that if you you can foresee your death about 2-4 seconds in advance, this can be a 100ish gold save in lesser bounty. This should be a RARE circumstance. If you do that 5-6 per game, then you are clearly taking way too many risks in the way that you play. Besides, it can be quite a hassle to retake weapons, and is clearly not worth the risk in my opinion. You are better off dying, make your opponent gain 50 gold (which represents about 4 seconds of creeping), than spending about 10 seconds (or very likely more) weakened because you must retake weapons. You creep less, are even more vulnerable to another death and clearly taking a lot of your time thinking of dying rather than be on the offensive or strategize globally. Example: You die, and you left your über weapon lying to save 150 on bounty. You respawn, but there is a crisis on hand: enemies are taking a CP. With your weapon, you could have gotten a double-kill, but since you left it somewhere else, you can't do anything and let the enemy take the CP. Thus, you lost a possible 600+ income go to waste AND a loss in territory, for a gain of 150 (balance wise). This situation happens way too often in my games (going defense just after respawn).

[RedNova]

Another take:

This strikes me as an abuse, not a gameplay option, even though the gain is small. Personnally I'd prefer bounty calculated on gold SPENT rather than actual slots value + chassis + ratio. Thus, a trader that upgrades creeps like a madman (and thus is extremely dangerous, probably more so than any other tank in the team), would have a very high bounty, instead of the paltry sum he is almost always worth (and even more so because traders seldom have a positive kill/death ratio).

[Prog]

If you do that 5-6 per game, then you are clearly taking way too many risks in the way that you play.

That's absolutly right, becouse its just a small mistake not to drop a item, but a huge mistake if you drop it and can't get it again. I use it approximatly in 1 out of 5 or 6 games. Its just a really small trick, but I like to play as good as possible, so every part of saving gold is welcome.

- This means that items not in your inventory do not count. Thus, when you are left with no slots availiable (and got cash to spare on extra weapons and items), you can customize your tank with no added cost when you die. For example, you play exploder, but opposing team is in siege. You drop explosives and go for a trader while they are stranded in their base. If they manage to break out, you revert exploder and retake your explosive. So in that situation, the fact that you bought the explosives will not impact your bounty when you die as a trader.

This is true, but i never see this in game, especially becouse a trade nearly never dies when the opponent is in siege.

This strikes me as an abuse, not a gameplay option, even though the gain is small.

I don't see it as an abuse, rather as a part of the game, like dodging rockets whith teleporter/invisibility, or AoW-Creeping in normal Wc3.


Mana Batteries or bought xp do not count to tankcost.

[kittyonmyfoot]

I've killed so many people by just a hair and dropping weapons would have most certainly gotten me killed. To me the decisions are that you would not stand and fight a loosing battle and that if you heavily out gunn them then you wouldn't need to drop your weapons. Ergo... if it's close and you have a chance... why would you drop your weapons?

Really the only time I item drop is if I'm a trader with decent inventory going into dangerous territory... or if I plan to do a complex trade and I want to reserve part of it until I complete the trade after going to the other side of the map... like going for factories and I'm all loaded up on salt and I need to get rid of water and wood that I've been collecting for revenue.

[Prog]

Ergo... if it's close and you have a chance... why would you drop your weapons?

I just quote myself:

Summary of points which show you when you should drop your weapon:

- You’ll be dead quite certainly
- No or only few creeps around you
- No or very small chance of killing your opponent
- You can reach your weapon easily and fast

This implies that it's not close. Of course noone should drop weapons when it's close, that's quite obvious. But it can be used in situations like: after a fight you escape with your heavy tank with 600 hp and reach your control point when you see an opponents heavy tank rocket flying to you.

[ancientsurvivor]

Isnt Indeterminismus a philosophical model ^^... Well the whole problem this is about is about someones micro, it is a good idea for people who have nimble fingers... but for the standart middleclass-players or the pest plants called noobs (hey weve been alle noobs once remember?) cant use that tactic in an apprpriate situation imo.