Milskidasith

   Your point is valid about it being acceptable sometimes. The problem is I can not count how many times  I have been paired with somone who keeps getting themselves killed but only suceding in baging the enemy half the time or less. After a few of those feeds our oponents catch on and just keep leading him into the towers, from there it's gg. Even if he finaly learns not to do it anymore they use towers tolower his health a few times and once he needs the crystal they continue buisness.  If there was only 2 lessons I could get my team mates to accept that would make many games go much more competitive it is these.

1. Do not fight the enemy EVER when they have the advantage. No mater the outcome  in the end you will loose more resourses than they do. ( this means no chasing into towers if you can't guarantee a kill and NEVER EVER if you are outnumbered)

2. If your 1 on 1 battle to the death with a DG failed and you were about the same strength or more at the start than him, DO NOT go for a re match under the same conditions. It will go worse the next time I assure you. This sounds obvious but I can not count how many times my partners are trying to fight the same oponent again after he just killed em and big surprise he wins again. See while your waiting 30 sec to respawn he is creeping exp and spending cash, he has not gotten weaker while you were gone he has goten even stronger.

If players would focuse on these two things every game would be a fairly close one, and I think we can all agree the games we enjoy the most and victorys we savor the most are the close ones.  Landslide losses are no fun at all for the loser and for me at least landslide wins are boring.

It also depends on your build.  I often play an AA Reg, so I'm very willing to exchange kills.  Gold is more valuable to my build than pretty much anybody else since its an equipment heavy build.  If I'm playing as say Oak, then I'm much less likely to exchange.

You also need to know a little bit about your opponent.  If you opponent has been using health potions, its really not a good idea try to kamikaze him because your calculations will probably be off.

Never kill an enemy if you are going to die after doing it.

I'm sorry, I thought getting X amount of gold and almost a level up worth of EXP is worth dying for. Considering the fact it gets me more than I would by farming creeps and since one DG on each side is out, it is a net benefit.

The entire guide was written so that people like you didn't spam "never kill if you are going to die afterwards" without giving any thought to the fact that, by getting the kill, you get a huge chunk of EXP and gold and the enemy gets nothing.

This is especially not true in a number of situations.  For example, if you are going to die to a tower, you should absolutely kill the enemy even if you are going to die.  Kill gold is worth more than the assist gold your opponent gets.

I try and aim for an 80% success rate.  If I successfully exchange 80% of the time, I'm happy.  If I die and my opponent doesnt die more than 20% of the time, then I'm unhappy with my play.

BTW, there is no assist gold when dead.

Sorry, it was a worthless and noncontributing joke as I essentially just copied and pasted that line from the OP.

On a more constructive note: still never go for a kill if you'll die in the attempt.  Ever.  I suppose I could read the OP and see what merits it actually DOES hold, but I'm still never going to die for a kill.  Or at all for that matter.  The most important tip for this game:  never die.  After all, how else do you expect to get the "deaths" reward at the end of the game?  Not by dieing!

Seriously though, just don't die.  The kills will better present themselves throughout the course of the game, with little to no risk of your own death.  Just play smart and be patient.  Going for that kill to death action will not win you the game.  I'd even be willing to stretch this to say that an X kills and ZERO deaths is substantially more demoralizing thing for the enemies to see when they hit tab than the feeling they might get knowing that even though they were killed by you, vengeance was theirs in the end.

You make money MUCH more quickly with successfully exchanged kills rather than a drive off.  Its not even close.  People who dont exchange if they get the chance are simply wasting gold.

I dont know about the XP, but if you are thinking gold, you will always have much more gold if you exchange a lot vs low kills and low deaths.  This is a very big deal if you play a money dependant build like I usually do.

Terrible advice if you are playing a build that is very money dependent (like say an AA reg).  The simple fact is that a kill generates so much more money than creep farming that you will get gold MUCH more quickly with exchange kills, while they will only get gold slightly more quickly.  The differential is greater in a killing vs driving off, as is the absolute amount of gold earned.

If you are playing something like an AA Reg and you are playing it safe and slow, you are hurting your team.  That build needs money to be really effective, and the fastest way to get money is to kill, even if it results in your own death.

Let me put some numbers to this.  On cataract, you get something like 100 gold for farming a wave of creeps.  So even if you only generate 1000 gold on a kill, thats still the same as killing 10 waves of creeps.  Even accounting for the fact that you wont be making money while dead and will probably miss some waves while dead, its still pretty clear that you will make MUCH more money with an exchanged kill.

Even if they get an assist on the kill (which is unlikely since they dont get assists after death), you are still going to come out ahead because while they are dead they werent farming either.  You would have needed them to be driven off for 4+ waves to make up the gold differential that you generate with a single exchanged kill even IF they get an assist somehow.  If they dont get an assist and instead forces of light/dark get it, then you would have needed to force them off for 10 waves to make that differential up.  That isnt going to happen.

First off, it isn't relevant. Secondly, big NOOOOOOOOO.

If you are at the point where chasing into the towers would kill you, you are going to have to run or die when the enemy comes back. And if he chases you, he will have enough HP to go through the towers. That's the situation. If you have enough health to chase and live, it's an obvious kill. If not... then you are going to want to trade deaths because it's better than having 20 seconds of creeping and being pushed away and possibly chased yourself. Not only that, but even the benefits if you don't get pushed are miniscule; even at the lowest level, you get 600 gold for a demigod kill. That's about as much as a creep wave with giants in it. Unless the enemy is buying giants at level 1, you aren't getting more gold for the kills, and the EXP is the same.

Yes, and then you are cannon fodder and you get a miniscule amount of creep gold and EXP. Really good strategy.

Contextual? Unless the enemy has giants at level 1, you are getting more gold and comparable EXP with creeping, AND you aren't a sitting duck for any other enemies. If you live at low health, a single enemy hurting you will get them a kill or push you away. So it's between death+ gold for you and death for the enemies, or no gold for you and maybe dying compared to the enemies just getting pushed.

So you say you are going to leave, and then post anecdotal evidence to support your position... nice. Hypocrite.

Also, Never Die is a great philosophy, but you don't seem to understand why people say it. It's because dying feeds the enemy gold, and it feeds them EXP. If you get a suicidal kill, they get neither. So it's more like getting pushed out of the lane, because the enemy doesn't get anything and you are out of the game for about 20~30 seconds. Hell, on big maps a suicidal kill is FASTER than going to the crystal to rest your weary DG.

The team with the gold and XP advantage will win the vast majority of the time.  Especially if they establish that advantage early (which is probably when most kill exchanging happens in my experience).  That makes trying to get a gold/XP advantage a very smart strategy.

Over half the time your team will have the flag anyway.  You either have the flag already or it was a 2v1 (which is why they were running) and your teammate will get the flag.  And if you are giving up gold/XP advantages to get time to damage a tower a bit more, well, I have to question that decision.

In conclusion, I dont see any rational way that you would want to exchange a small amount of warscore (on average) and the possibility of doing a small amount of damage to a tower for 1k+ gold and a solid chunk of XP.  That exchange you are suggesting doesnt make any sense to me.

Point 2 is wrong if one of the assists is the guy who is now dead.  The people getting the assists would have to be 2 nondead people, which is a rare situation.  Plus it doesnt even out since I think you will come out ahead in terms of XP, while the gold is even (somebody can check me on this, not sure).

What you are missing from all your "you need to be alive to have the possibility of doing something else" is that by killing your opponent you also remove them from play for 30 seconds or so when they probably could have been back in play much more quickly by simply doing a quick crystal run and tp back.

Not to be mean, but you should be here to change your opinion.  The whole reason to read strategies is to change your mind about how to play the game.  Why are you reading a strategy board if its not your intention to change your reasoning?

We have suggested detailed ways in which kill exchanging is better.  All you have done is repeat the EXTREMELY simplified saying that "I'm going to try hard not to die" and not stated how your strategy will work better than intelligently pursuing exchanged kills.  The "dont die no matter what" is great advice for a newbie.  A newbie shouldnt be trying to kill exchange.  But you have to accept that once you get better at the game, some of that simplified advice for newbies is going to get replaced by more nuanced ideas.

Ok...

I dont know exactly who you are ranting about since those definitely arent my stats (I have quite a few more games than that - over 100 recorded according to the website).  But suffice it to say that you are being pretty silly here.

Trying to use game stats to prove anything is just stupid at this point in time.  Not only are there so many questions about games recording and whatnot (in short, stats DO lie in DG right now, and they lie badly), but there is also a big difference between somebody who plays randoms and somebody who plays premades.  You play premades against random teams.  Good for you.  Nothing wrong with that.  But OF COURSE you are going to have a better record than people who play PUGs.  And beating a PUG with a premade certainly doesnt prove anything about your skill.

Someday we will probably have tournaments and stats that mean something (maybe even with this next patch).  At that point you can run around and be all proud about your record.  Hopefully by then you will have found some maturity.  But, until that happens, please dont pollute the boards with garbage like this.  If you want to argue a point, fine, but at least try and make a rational argument.

Thanks.