Have you ever thought about what factors lead to throwing a game that should be easily won? I'm CarlTheLlama, #1 ranked Hero League support main, and I've been a part of more than my fair share of throws. Here's my advice on how to more consistently secure games that you take a lead in, and stay in the lead.
Good stats do not replace good play
Having better stats than the enemy team, or even being up a talent, allows you to be more confident in making fundamentally sound plays, but it doesn't allow you to create any type of play you want. For example, being up 3 levels (including a talent) as Tyrande isn't going to let me make a solo pick on a target, even though Tyrande is phenomenally good at making picks with a partner. While in general, making picks is one of her strong points, it's contingent on having a buddy to help with the damage, and a stat lead doesn't change that. What's more, from my opponent's perspective, a squishy like me who doesn't have tons of 1v1 potential is going to be their best chance to get a kill and start their comeback. Even if my play somehow is successful, that means the game is so far gone that even the best opportunity given to my opponent isn't something they could capitalize on- the game was already won and my play didn't help that at all. In the future, it's going to make me misjudge situations, be overconfident, and potentially throw that game, or perhaps generally play too riskily in an even game. Whatever the stat situation is, you have to remember what your character is fundamentally designed to do, both their strengths and weaknesses, and compare that to the enemy hero.
The same applies on a team level. Trying to take a boss against a barrel Chen isn't going to work even with a talent on him and his team not there. Splitting up for 3 Towers of Doom altars against a heavy pick team isn't going to create good scenarios for you even with 3 levels of stats. Fighting against a team with heavy AOE at a merc camp is going to lead to you being wombo comboed. Being ahead doesn't mean you can do anything you want, it means you can freely do what you're primarily equipped for, and do more of the things you're partially equipped for.
Hand in hand with the idea of continuing the type of plays that got you ahead in the first place, is the idea of thinking about what types of plays your opponents want to do to win a fight despite being down levels, and avoiding those. The last paragraph dips into that idea at the level of what kits are aimed at, but you also have to consider how the object of dispute (e.g. a fort, merc camp, or boss) affects the equation. When winning, you need to press your advantages and in so doing expose yourself to some level of risk, but it needs to be a risk you've calculated.
Overweening pride is when your eyes get larger than your stomach, and you make a play thinking only of what you will get from it, and not thinking about the additional risk involved. It's when Zeratul used the same energies as the Overmind to slay Zasz the cerebrate in Brood War, and in so doing telepathically connected with the Overmind, revealing the secret location of Aiur because he didn't stop to think about what would happen when those energies connected. That's a lore perfect analogy for what it is to throw. Just as such a mistake resulted in the exodus and mass slaughter of Zeratul's people, taking a boss when the enemy team can react to it is a mistake of overweening pride that gives an enemy team the best of opportunities to make a huge comeback play. The key to staying ahead is denying the enemy's chances to come back. When you're ahead is the time to be the most careful.
The Throws. Reporting Zera for dooming Aiur.
Even if by some miracle no one gets CCed by the boss while the enemy team is watching, and by good play you finish the boss as the enemy team gets to the point, and by another miracle you have the sustain to have everyone at full HP and full enough mana when the fight starts, you're going to be missing 1-2 regular cooldowns from most of the team members, used on the boss. You might as well have 2-3 heroics not be up for 10 seconds after the fight starts, which compounds with the fact the frontline/backline positions are reversed, putting your entire team out of position. Remember, that's a best case scenario that assumes the enemy team engages after you've finished boss, not during it when your focus is split, or any number of other problems that could arise. If you can win that fight, you can certainly be winning fights elsewhere, and those are the options you should be pursuing.
Boss fights aren't the only place where your eyes get bigger than your stomach and you don't respect your opponent. Overstaying is another great example of telling yourself "Oh, I want to get this building/kill" without thinking "my health/mana is low, and it's a long way to base." The wrong hero tanking fort/keep debuffs can be unexpectedly devastating, and the same goes for letting multiple people/squishy targets get hit by core- it's a seemingly small thing that you don't think about until too late because you're confident from your stat lead. The key to preventing yourself from being a victim of your own overweening pride is to ask yourself "If I was in their position, is this the type of situation I'd be looking for to make a comeback play?" Don't just avoid those scenarios when they happen, anticipate when you're going to reach that point before you begin the play, and have your escape plan prepared.
The Law of Toast
Do you know how to make toast? Most of you probably answered yes, but most of you are probably wrong. To get the butter to soak into toast, it has to reach to bread immediately upon exiting the toaster, and then in the few seconds while the toast is still hot, you have to eat the toast. A seemingly simple process, but it gets screwed up if you try to do it with more than one piece of bread at a time. If you try to do it with two, the butter won't soak in, the toast will be cooled off by the time you eat it, you'll burn the toast in the toaster, etc. At least, that's the case with every toaster I've ever worked with. The moral of the story is: even for simple jobs, if you want things done right, you have to do one thing at a time in order to meet the real-time requirements necessary for quality control.
Doing 2 merc camps is fine and good, so long as you do them as a group, one at a time. Otherwise, you're exposing yourself to getting picked, losing heroic cooldowns before the teamfight, or at the very least donating a camp to your enemies. Even if they don't punish you, you'll incur more damage from the camp than you need to, globes won't replenish resources for as many people (or stack for heroes that use stacks), and you gain absolutely nothing over doing them one at a time as a group.This is more noticeable the bigger the object you're contesting is. For example, you know not to take both bosses on Cursed Hollow at once, because you're incurring more damage than you should and not close enough to help each other when the enemy team shows up. The same principle is true whenever you're split and the enemy team is grouped, it's just less noticeable and less likely to be exploited- but the better your opponents are the more they can exploit an opening, no matter how small. When your team is trying to do 2 things at once, like push and take boss, or push 2 different lanes, or take a merc camp during a map objective, both groups are more likely to fail against any organized retaliation by the enemy team, since being down a member of your team is going to hurt more than almost any level/talent advantage is going to help. So far this point has been pretty basic, but the subtle thing about the law of toast is the element of time.
If you find your team pushing 2 different lanes, obviously you want to tell people to regroup and focus one place. However, it's not when the person(s) mount up and head back to top lane that you're firepower is refocused, it's not the moment when they meet back up with the group that you're refocused, it's the moment when you have all cooldowns ready, and the HP/mana necessary to operate your hero back up that you're refocused, and by the time that happens, the boss pushing down top lane might not have enough life left for you to force the fight you wanted. The toast is ruined, and trying to force a fight is not going to end the way you wanted. Of course, whether that was a critical factor or not all depends on how much power your level lead is giving you. The important thing is that you recognize that opportunities don't last forever, and what was a great play 5 seconds ago is a terrible one now. Doing things as a group allows you to move more quickly, and fit through more of those windows of opportunity, but it's not going to expand the window itself. Be focused enough that you do one job at a time, and get anyone back on board who falls off, but don't be so tunneled that you miss the expiration date and follow through after an unforeseen delay that changes the critical factors in a play.
So basically, the way to keep yourself from throwing games is check yourself before you wreck yourself by asking yourself 3 basic questions: "
1. What types of fights is our comp favored to win?" Short and bursty or long and kitey, sieges or picks in rotation, in tight areas or in large ones, etc.
2. "How much risk is involved in taking X option, and what specifically creates that risk?" If your opponents focus and make one good play that is perfect (despite whatever they've been doing up to this point) will the weakness you incur from the option you've chosen be enough to let them beat you?
3. "Is everyone on the same page, or is there any human error affecting the planned timetable?" The human aspect of strategy, both in your team and yourself can often create problems that seem minimal, but throwing off the timing of a plan can have major repercussions, which means you have to re-assess.
Rather than making this a complicated activity that you pain over, you want to internalize these questions to the point that they become a habit that is quick, efficient, and doesn't take much mental energy so you can still focus on other factors in the game. The best way is to tie these questions to a concise idea that has a catchy rhythm, like "Good stats don't replace good play," or a phrase that holds significant emotional meaning to you because of things you associate with it (such as the plot of Starcraft), like "Overweening Pride," or something quirky or humorous like "the law of toast." As long as it helps you compress more complex ideas into a simple category, you'll find that with repetition you develop both speed and thoroughness in best practices that will keep you from throwing games.
Thanks for your time, I know you could have been pwning noobs with it instead ;)
This is a doc that I keep current of my articles, videos, stream, social media, and anything else I think is of interest to the HotS community. https://goo.gl/eYVNEq