17 replies to this topic

[JigglyMoobs]

For mechwarriors who are looking to up their game, I highly recommend the following reading:

http://en.wikipedia.org/wiki/OODA_loop

http://www.ejectejec...ves/000172.html

http://www.forbes.co...-the-ooda-loop/

Forty years ago, a little known US Air Force pilot named John Boyd elucidated a set of insights and principles that have fundamentally changed how modern strategists think about conflict and competition.

These articles tell the story of what OODA is, how Boyd arrived at his revelations, and the impact of OODA on every aspect of our modern world. They are long, but well worth your time. Not only will OODA help you win at MWO, it will give you a new angle from which to look at the world, and inspire thinking and practice that will help you throughout your personal and professional lives. I highly recommend it!

Edited by JigglyMoobs, 02 August 2014 - 07:09 PM.

[JigglyMoobs]

By the way, the development of this game, the bazillion updates, the constant tweaks and nerf? It loosely follows a software development strategy called agile software development. Guess where that came from?

http://blog.practica...d-software.html

Yep. The same philosophy that led to the development of the American F-16 is what's making this game today.

Edited by JigglyMoobs, 02 August 2014 - 07:16 PM.

[Maggiman]

I really don't think this applies here. OODA (as far as i understand, i might be truly wrong here) is about constant small adjustments based on observations. The thing is, the loop needs to run trough quickly which is in my opinion the opposite what happens here. Example:

PPC too strong (for whatever reason) -> 5% more heat in the next patch.
Still to strong? 5% more heat *next* patch.

Reality: PPC too strong -> Hard nerf (Though i think it happened in 2 patches last time?) and another look in a few months.

I believe they actually stated(Again, my fault if i remember incorrectly) the have a loop which they go trough for every weapon (small laser? ok..medium laser? ok.. and so on) which actually takes a long time. Which means pressing concerns are ignored for a set while. This is in my opinion exactly counter to OODA. (They got Observe, but in the wrong way)

Of course, i don't mean this as a qualifier whats best, pgi's way or OODA or whatever (I think the article shoehorns the term onto software development a bit anyway)

Edited by Maggiman, 02 August 2014 - 07:54 PM.

[JigglyMoobs]

I don't think that's quite right. My understanding is that OODA is more about the speed of decision making rather than the exact decision.

For example if pgi looked at a bunch of variables that something isn't working and decided to do "something" quick, that would be working with OODA principles.

It doesn't matter whether that something is big or small. OODA says what's more important is that they gather their data, look where they are and do something quick, and immediately after that, start gathering data etc again as fast as they can so that they can adjust again. This "something" can be a small change or a major Nerf.

If pgi decided instead that they will change something and not touch it again for a year no matter what, then it would be counter to OODA principles.

Also OODA is a very general strategy. How well and to what degree its executed in particular instances can of course vary. Almost nobody does it slavishly, but many organization utilize it to some degree.

Edited by JigglyMoobs, 02 August 2014 - 11:00 PM.

[YueFei]

The swordsman analogy from the ejectejecteject site is interesting. It may interest you to know that European Martial Arts masters left instruction manuals, stressing the seizing of initiative... and then holding the initiative throughout, being "ahead of the opponent's decision loop", as it were.

What was supposed to keep the attacker safe was geometry, closing off his opponent's line of attack when launching his own. You couldn't afford to hesitate. Closing in decisively with an enemy prevented him from faking you out or changing up his swing. That way, even if he did, he wouldn't have enough space and time to wind up a swing that could hurt you.

In order to make decisions, you have to at least have a set of choices from which to pick.

By developing those techniques, swordsmen would have the confidence to step in and engage, knowing that the technique would keep them safe.

An example of a bad technique would be to just thrust directly at an opponent without getting some kind of opposition on his blade. You might stab him, but there's nothing preventing him from stabbing you simultaneously for a mutual kill. To the European masters, that was an unacceptable outcome. Forget modern sport fencing for a moment, where they have "right of way" and the first to touch wins the point. That's a game, not life-or-death combat. Just because you mortally stab someone doesn't mean they immediately stop acting, and they can still act long enough to mortally stab you in return.

Anyways, the point of me bringing this all up is that, in MWO, there are a lot of players paralyzed by "fear" and indecision. I put "fear" in quotes because almost nobody is actually scared of dying in this game. It's more that they're trying to not lose, they don't wanna get gibbed and be the reason that their team lost the match. But that's because nobody ever taught them techniques they can use to favorably engage a particular type of opponent.

If the only move a swordsman ever learned was to thrust directly at a foe, you'd see a lot more hesitation. He'd constantly be trying to dance around out of range of his opponent, not seizing initiative. If he's fighting someone also equally ignorant, they'll be playing sword-tag nicking each other all day until someone finally dies of blood loss from a dozen small cuts and puncture wounds. But if he fights someone that knows the other techniques... he won't be expecting the sudden ferocity and aggression coming from the enemy, who will suddenly be upon him, and the fight will be over fast.

My point behind all this is that, if you want more MWO players to become better pilots, they have to be taught those techniques that will allow them to engage the enemy favorably and help the team win. Otherwise, they can learn about OODA all they want, but in the end if the choices available to them are bad, deciding quickly still isn't going to save them. There's a difference between "bad" choices and "sub-optimal" ones. You can make "sub-optimal" decisions quickly and do well. But making "bad" decisions quickly just gets you killed.

Edited by YueFei, 02 August 2014 - 11:34 PM.

[JigglyMoobs]

Oh absolutely Yue Fei! Efficient application of OODA in a fast moving situation requires that the practitioners develop a useful repertoire of skills, tactics or stratagems that they can execute. This gives them the "agility" necessary to act.

In the case of MWO, that's everything from torso twisting and jump sniping on the level of an individual pilot to group strategies like the counter clockwise circle of death on River City. By practicing the elements of this repertoire as an individual and a team, you gain the agility to rapidly put your decisions into practice.

[Maggiman]

While i would say, that pgi is too slow for that kind of development i heartily agree with the bad decision making of pilots.
Exactly the fast reasessment of the situation is missing.

[Appogee]

"Agile Development". Increasingly being used as an excuse for "release it half-assed and patch it later".

[YueFei]

JigglyMoobs, on 03 August 2014 - 01:42 AM, said:

Oh absolutely Yue Fei! Efficient application of OODA in a fast moving situation requires that the practitioners develop a useful repertoire of skills, tactics or stratagems that they can execute. This gives them the "agility" necessary to act.

In the case of MWO, that's everything from torso twisting and jump sniping on the level of an individual pilot to group strategies like the counter clockwise circle of death on River City. By practicing the elements of this repertoire as an individual and a team, you gain the agility to rapidly put your decisions into practice.

Yeah. One of the most impressive things I've seen good pilots do is to actually *not* take the shot opportunity when an opponent suddenly appears, and instead they'll twist / turn / jump and shield first! Usually they do this when they realize the enemy is looking at them and also ready to shoot, and it'll be an even trade. So instead they'll absorb the hit on an arm, and then maneuver to keep the enemy in their sights, and *then* hit the enemy. Often they'll do this while maneuvering into a superior position. Even if the enemy also shields, it becomes an even trade, minor damage to both parties, but the superior pilot is getting into a position from which he'll have cover and the enemy won't, so he'll have the initiative to attack.

By absorbing a 20 or 30 damage hit on their arm, it's as if they added 30 armor to their LT/CT/RT (where-ever the shot would've hit if they didn't shield). That's HUGE. Who wouldn't want to be able to add 20 to 30 armor points to their torso?

The other aspect is the team aspect. If two teams are at opposite sides of a ridgeline, but one team just sits there taking potshots of opportunity, and the other team crests the ridgeline in coordinated attacks... even if both teams have equal accuracy, the team that crests the ridgeline in coordinated attacks will win. Let's say Team A is the timid team, and Team B is the one who seizes initiative. Team B crests together and fires at the same 2 or 3 mechs of Team A. Team A's return fire falls upon the first 2 or 3 mechs on Team B that crested. But on the next wave of attacks, Team B cycles their hurt mechs into the back, and then crests together again and shoots the same 2 or 3 injured mechs on Team A. Meanwhile Team B's hurt mechs fire from behind the shoulders of healthier teammates. Eventually the 2 or 3 mechs on Team A that are being hammered repeatedly will die. Meanwhile nobody on Team A seizes the initiative to crest that ridgeline and deliberately try to finish off Team B's hurt mechs. The process repeats until all of Team A is dead, and many mechs on Team B are damaged, but alive and kicking with weapons intact.

Sometimes in PUGs you realize you're the only one breaking from cover to fight. You're essentially on Team A with a bunch of timid players. From the point-of-view of the other players on Team A, they think you're an idiot. They don't realize that their timidity is actually hurting the chances of the team winning, and thus hurting their chances of surviving the match. As mechs on Team A get focus-fired and killed 1 by 1, from the PoV of the mechs that survive toward the later half of the match, each player sees teammates falling but he himself is still alive. He attributes this to him being smart and staying behind cover, when it fact it's because of the team being passive that they aren't seizing chances to attack and finish off hurt mechs on Team B..... mechs on Team B that continue to shoot at and kill mechs on Team A.

[JigglyMoobs]

Yep, you see this all the time on PUGs. People are way too timid and they keep their heads down, basically denying themselves every aspect of OODA at the team level, and then at the individual player level, they fall back to instinctual actions only once they are in a fight instead of making conscientious decisions to minimize their damage intake and utilize available terrain.

On the other hand, when you are running with a competitive group of players, everyone is practicing OODA (knowingly or unknowingly) all the time. Lights are constantly on the lookout, reporting back enemy positions, de facto group leads are always assessing and reassessing based on latest available info, the team coheres around those decisions and put them into action efficiently, and then individual players are taking smart tactical decisions for themselves within constraints of the overall tactical plan (which is itself very fluid).

The difference is night and day. I was rolling with some competitive people the other night, and it literally takes them about 5 seconds at the start of the match before everybody agrees on a plan and starts executing it. This leads directly to <3 min 12-0 rolls of non-competitive groups (4 man, 10 man, 12 man doesn't matter) because those guys get hit by the group enmass when they are scattered, mentally unprepared, have no idea where our team was and had no plan of action. Now, there were a few instances when they run into things they don't expect, and instead of being paralyzed they change their mind *like that* and everybody will instantly switch to executing plan B.

I'm still having trouble with it because I have trouble integrating my tactical approaches at the individual level with what the group is doing. I'm still adding to my moves if you will and it's a slightly different repertoire than what you use as solo pilots.

You should come drop with us Yue Fei! These days it's a lot more fun than running solo.

PS - if I'm not around try checking out JagerXII, MWOPRO on twitch and seeing if anybody want to group with you. Also, you can message SLDF Deathlyeyes, SLDF Xavier, Pepsi One, SLDF Medi0cre or other SLDF guys if you want to hang out with our clan.

Edited by JigglyMoobs, 03 August 2014 - 02:04 PM.

[Tim East]

Driving a light can be a lot like piloting a fighter plane, I suspect. Erich Hartmann's see-decide-attack-reverse philosophy will serve you well. See the enemy, decide how best to approach, make the attack, and disengage to re-evaluate the situation. I tend to play with very close-range weapons on my mechs, so I'll often look for a gap in the facings of the enemy death ball. Much of the time, people only look toward your team's death ball, so you can run behind an enemy and rip off his back armor in a single pass, or, if you feel ballsy and suicidal, you can glue yourself to his back and try for a kill. Usually it's better to disengage entirely or strafe the enemy formation though than picking a single target to focus on as a light though. I've died more from failing to disengage and re-evaluate than anything else.

[YueFei]

JigglyMoobs, on 03 August 2014 - 02:02 PM, said:

Yep, you see this all the time on PUGs. People are way too timid and they keep their heads down, basically denying themselves every aspect of OODA at the team level, and then at the individual player level, they fall back to instinctual actions only once they are in a fight instead of making conscientious decisions to minimize their damage intake and utilize available terrain.

On the other hand, when you are running with a competitive group of players, everyone is practicing OODA (knowingly or unknowingly) all the time. Lights are constantly on the lookout, reporting back enemy positions, de facto group leads are always assessing and reassessing based on latest available info, the team coheres around those decisions and put them into action efficiently, and then individual players are taking smart tactical decisions for themselves within constraints of the overall tactical plan (which is itself very fluid).

The difference is night and day. I was rolling with some competitive people the other night, and it literally takes them about 5 seconds at the start of the match before everybody agrees on a plan and starts executing it. This leads directly to <3 min 12-0 rolls of non-competitive groups (4 man, 10 man, 12 man doesn't matter) because those guys get hit by the group enmass when they are scattered, mentally unprepared, have no idea where our team was and had no plan of action. Now, there were a few instances when they run into things they don't expect, and instead of being paralyzed they change their mind *like that* and everybody will instantly switch to executing plan B.

I'm still having trouble with it because I have trouble integrating my tactical approaches at the individual level with what the group is doing. I'm still adding to my moves if you will and it's a slightly different repertoire than what you use as solo pilots.

You should come drop with us Yue Fei! These days it's a lot more fun than running solo.

PS - if I'm not around try checking out JagerXII, MWOPRO on twitch and seeing if anybody want to group with you. Also, you can message SLDF Deathlyeyes, SLDF Xavier, Pepsi One, SLDF Medi0cre or other SLDF guys if you want to hang out with our clan.

Yeah, group drops are a different beast altogether! My first few forays ended embarrassingly for me, I ended up getting overrun and killed early. Good groups are much more aggressive, especially if they know somebody is isolated or has minimal support. Whereas PUGs hesitate so I have more time to withdraw, in group play, it seems like 4 or 5 enemies are rounding on me at the same time.

You spoke of crushing non-competitive groups who were scattered and not prepared? I was one of those guys getting squished!

A big part of it was my fault for positioning myself so that my own teammates behind me did not have good angles to lay down fire support for me. I need to work more on placing myself so that if the enemy decides to take me, they lose far more than they gain. Like taking an enemy's Knight but losing your Queen and Rook in the process. I have a feeling that I need to communicate better, so teammates know I'm going to bound forward and they can setup their firing lines in advance.

If I say nothing until that moment when I mutter on TS "I'm in trouble".... it's already too late for my teammates to setup and start shooting.

I just realized I did play with one of the SLDF guys a few weeks ago, though I didn't know he was SLDF. "John 8osk" I think it was. There were probably other SLDF guys in there, too. These guys were doing so well, they made it easy for me to do well also. I had no problem closing on the enemy in my brawler HBK, because they were shooting them in the face so hard they were afraid to peek out.

And when I did go in for the close range fight, these guys were right there pouring fire into the enemy before I could get myself into too much danger.

Edited by YueFei, 05 August 2014 - 08:36 PM.

[Carlos Vinson]

For those interested I recommend

Boyd: The Fighter Pilot Who Changed the Art of War
Robert Coram

&

The Mind of War: John Boyd and American Security
Grant T. Hammond

Lot of good stuff on Boyd out there. Really fascinating person and Important ideas

[JigglyMoobs]

Carlos Vinson, on 06 August 2014 - 06:26 AM, said:

For those interested I recommend

Boyd: The Fighter Pilot Who Changed the Art of War
Robert Coram

&

The Mind of War: John Boyd and American Security
Grant T. Hammond

Lot of good stuff on Boyd out there. Really fascinating person and Important ideas

Great stuff.

[YueFei]

Hey Jiggly, I just had an idea. Reading about Boyd's "Flat-Plating the Bird" maneuver got me thinking.

Is it possible to make yourself nigh-unhittable by travel-time weapons if you have a slope with a taller structure on top of it, facing toward the enemy?

For example, on Crimson Straits, using the saddle. There's a smaller structure on the top of that saddle. If you jump laterally from the base of the saddle, so that you're exposed at the crest of it but moving laterally toward the building so that you'll slide behind it ASAP, anyone trying to hit you with travel-time weapons will have to lead the shot.... perhaps leading it at where the building is. If the building wasn't there, they could lead the shot to hit you, but with the building in the way, they can't. And since your own projectilesl don't inherit your velocity vector, you can still fire just before you pass behind the building, and land shots on the enemy.

It might be even more do-able now that PPC velocity has been reduced so drastically, forcing people to lead their shot further ahead.

Basically, you want to use the vertical dimension to crest some kind of hill/ridgeline, and simultaneously you're using the lateral dimension so that as soon as you clear and shoot, you're gonna fly right behind a building/structure/comm tower that's at the top of that hill/ridgeline.

Heck, you might even be able to use the hill itself. Instead of using the right-hand-side of a hill to shoot around with a "right-handed" mech, you creep below in cover at the left-hand-side, run up, jump laterally towards your right, shoot at the crest, and fly behind the hill.
(EDIT)
Wait, no, I was confused about the direction. Using the left-hand-side of the hill would increase your exposure time if your mech is right-handed because you need more clearance to shoot or your right-hand-side weapons will clip the hill. So actually you'd use the right-hand-side of the hill, but you'd jump laterally towards your left, moving at full speed, peeking out with just the right shoulder and right arm, taking the shot, and landing. Disadvantage is that you won't be able to use your left-side as a shield, so if you don't pull it off perfectly you probably get hit on your "sword" side. Unless you jump backward so you can shoot with the right and shield with the left? But if you move backward towards the left, you won't move as quickly, which means enemies don't need to lead the shot as far ahead, so they might be able to hit you instead of being forced to aim into the hillside.

I hope I'm explaining my idea properly, it'd be easier to explain with pictures...

Is this kind of maneuver possible? Can you guys test it out? Maybe it's possible only with certain mech geometries, jump speed, and land speed?

Edited by YueFei, 15 August 2014 - 06:10 PM.

[JigglyMoobs]

I think I kinda know what you mean Yue Fei. Will have to experiment with it.

[1453 R]

“In war, haste trumps cleverness. Strike quickly and you will end it quickly.”

Thanks for the links, Moobs. I deeply enjoyed the EEE article, and it’s always helpful to be able to put something your gut knows but your brain doesn’t into conscious words. Also, this thread has made me absolutely terrified/little-boy-eager to see what a trained F22 pilot could do in an Ember in this game. Talk about your 12/0 sweeps…

Anyways. The notion of seizing and retaining the initiative has been talked to death in this forum, but here is a shining golden example of why it works. Here’s the mechanics behind why early decisive action wins games, why a bad plan executed quickly and with gusto always trumps a good plan executed anemically by half a team four minutes into a match. Here’s why no competitive team camps for more than a few minutes outside extremely unusual circumstances, and on a slightly cheekier note why the Timber Wolf is so damned good (it’s clearly the F-15 Eagle of MWO at the moment).

Applying this precept team-wide, or even with a few brave souls willing to follow a decisive lancemaster’s lead, wins games. Your own personal butt may end up getting blown off, but even the dumbest and blindest of puglies can figure out they’ve got a leg up when the score’s 4-1 your way and Get The Memo. I’ve been trying to figure out, in recent drops of mine, how to better convince my puglies to bring death and ruination to the enemy instead of inviting it upon themselves. Generally that means leading the charge even if your machine sucks at it and hoping against hope that the guys behind you follow your lead.

Nobody wants to be the First Guy In, because that guy gets messed up…but if you volunteer for First Guy In, and you can drag even one or two other guys with you, you’ll be amazed how many more games you start winning. Case in point: I’ve given up trying to keep my team out of the center/Gladiator Ring/Mosh Pit/Pug Zapper in Mordor and have switched tactics to: blitz that sucker as fast as I can, and don’t stop until I’m in position by the enemy’s entrance. Push hard and aggressively into the center ring, drag your team with you by word, by deed, and by raw force of will, and your win rate on that godforsaken pit will triple. Mine did, anyways.

It’s not just about following OODA and trying to get inside the other guy’s think loop. It’s about doing everything you can to not let yourself fall into the trap of paralysis and demoralization a faster-thinking enemy puts you in. Even if they’re ahead of your loop, you can still get lucky and make the right choice in the right moment by sheer dumb luck and reverse the stakes – but only if you keep acting. Do your damnedest to keep up, soldier on and refuse to let the other guy faze you, and even if you’ve lost the initiative, you’re making the other guy work damn hard to keep it.

Better than him getting to seize it and keep it for free, and use it to club you about the head and shoulders until you’re too concussed to see straight and pitch over backwards.

Or, TL;DR…STOP HIDING BEHIND COWARD’S ROCK AND FIGHT. YOU MAY LOSE IF YOU DO, BUT YOU WILL LOSE IF YOU DON’T.

[JigglyMoobs]

Another thread reminded me of this one we had over a year ago.

Still damn important today.