@theskeindhu said in Russian Openings and AA Online:
Re: All the Russian Openings for Beginners
I took a hiatus and recently returned. I can’t win at all as the Allies. The most recent game I went conservative and stacked everything first turn on West Russia. Germany ignored the UK fleet and brought everything in to West Russia. They barely won and took Moscow round 3. I’m not a statistics guy, but is this a high probability attack? I lost 3 infantry opening attack, both AA Guns and every man was stacked West Russia.
@theskeindhu said in Russian Openings and AA Online:
. . . been stacking West Russia turn 1 with everything minus fighters which obviously can’t land there first turn. 1 to Egypt, 1 to Szechwan along with one infantry. This has been my tactic and the most consistently successful, since I got tired of gambling a strafe on Ukraine and getting major League screwed enough times, including losing a fighter first round that I decided it wasn’t worth placing the entire fate of the game on round 1 luck. Germany has about 20% success rate rushing a West Russia against my stack R1. This, IMO is because AA isn’t working as intended, a known glitch. The odds improve if I have 6 dice rolling for 1’s and actually take out aircraft.
Not a stats guy, not an issue. AACalc :sunglasses:
http://calc.axisandallies.org/?mustland=0&abortratio=0&saveunits=0&strafeunits=0&aInf=12&aArt=3&aArm=4&aFig=1&aBom=&aTra=&aSub=&aDes=&aCru=&aCar=&aBat=&adBat=&dInf=3&dArt=1&dArm=1&dFig=&dBom=&dTra=&dSub=&dDes=&dCru=&dCar=&dBat=&ddBat=&ool_att=Bat-Inf-Art-AArt-Arm-Sub-SSub-Des-Fig-JFig-Cru-Bom-HBom-Car-dBat-Tra&ool_def=Bat-Inf-Art-AArt-Arm-Bom-HBom-Sub-SSub-Des-Car-Cru-Fig-JFig-dBat-Tra&battle=Run&rounds=&reps=10000&luck=pure&ruleset=AA1942&territory=&round=1&pbem=
AACalc’s rounding and adding routines return odd results sometimes, and AFAIK you can’t assign hits to AA guns with it. But it does have the advantage of running 10,000, so your outcomes are probably going to be within 0.5% . . . ish?
https://www.aatoolkit.com/conflict
Only runs 1000 times so you get much larger swings on reports. Like 8% swings I wouldn’t consider unusual at all. But you can assign hits to AA guns so there’s that.
Anyways first link, I assume you hit W Rus with 1 fighter. You specified in later post you’re sending another fighter to Egypt. So actually this whole recent thread discussion about G1 with bomber against Egypt, with nobody assuming USSR has a fighter in Egypt (because mostly the replies are to Quintin, not you, the OP), you already headed off. You add a USSR fighter at Egypt, yeah, that stuff changes. So good on you there.
But returning again to OP, you’re asking about the G1 counter.
Let’s take aacalc’s 90%+ reported case. I won’t get into mathematical modeling and stuff, but generally speaking if you have a lot of dice and 85% an attack is “safe”.
But you say aacalc reports 100% Allied win? Look at the breakdown. Within those outcomes are different levels of USSR surviving. Sometimes USSR takes no losses, sometimes it takes a chunk of losses.
AAcalc has that rounding stuff so you have to watch out on outcomes. But in this case, you look on the chart, take 9 inf 3 art 4 tank as your “baseline”. When you do your all-in against West Russia (minus the fighter headed towards Egypt later), it’s not just you expect to win. You expect to win, and 90% of the time you expect at least 9 inf 3 art 4 tank to survive. Then you can move in two antiaircraft guns.
Mmm? Make sense?
Now we look at Germany’s counter. 6 inf 4 tank 4 fighter 1 bomber is the “greed” counter; Germany wants to mess West Russia up hard but they’re keeping back two fighters to hit the UK battleship+. But if Germany really wants it, they throw in two more fighters.
(EDIT - 18 January 2021, when writing a later post and reading up, I noticed I’d omitted Germany would also have an artillery. Pretty big miss for me; I included it in my projections in my 18 January post later in this thread. Preceding description and following numbers not accurate, apologies for the mistake.)
Using https://www.aatoolkit.com/conflict (selecting for 1942 and changing “take this unit last” to have USSR AAA destroyed first), running a few times (as it only runs 1000 times), the “greed” opening ranges from 14-24%, but let’s say 14%. Not great.
But when you don’t do the Germany greed opening, it goes to 44%. A lot better for Axis.
You can sort of play around with the numbers there, but it works out to something like this.
You do W Rus only open with fighter to Egypt, you have about a 90% to get forces that in turn only offer Germany a 14% on a “greed” counter. But if Germany goes all-in it’s 44% if you’re low-balling the USSR projection of USSR survivors of the R1 opening attack.
But then, there’s a 10% chance you don’t have survivors in that safety margin. If you play with the numbers, what does that work out to with the non-greed counter? Something like 66% Germany breaks West Russia. Not bad.
You can run projections on what others have advised, the W Rus / all-in-at-Ukr open. But I expect (without bothering to run the calculations), things are maybe going to be a bit soft.
I gave aacalc links in my first post in this thread. But you’ll notice my projections were, as I wrote, “Germany can hit W Rus on G1 with 3 inf 2 tnk 3 fighter”. This is the greed counter. If Germany does NOT run the greed counter, then W Rus/Ukr still isn’t safe.
Sure, there’s compensations. If UK battleship survives, there’s all sorts of fun shenanigans. But the model is now Allies can never land fighters on West Russia because it’s always traded, and USSR’s probably choked off on income, plus I expect Germany will retain control of Karelia.
I’m not saying it’s this big awesome obvious easy Axis counter. It’s fraught with peril, the Axis do one thing, the Allies counter, the Axis counter the Allies counter, the Allies counter the Axis counter-counter . . . right.
As I wrote in that other recent thread, you just have to have strong fundamentals. I can quote percentages, but what do those percentages really mean? What are the positions that can develop? If things don’t actually occur within the projected acceptable range of results, what are the contingencies?
And as I wrote there, you just won’t get a “safe” line. It’s not that I’m saying abstractly it’s impossible to have “safe” lines. But it is a dice game. And further, it’s my belief that 1942 Second Edition (I know, we’re talking about 1942 Online) was designed exactly so you have these different dice outcomes that happen in different games. The different dice outcomes occur, then how players deal with those dice outcomes is how the game is played. If an attack is 85% safe, it’s also 15% UNSAFE. And what happens when that 15% occurs? If you take all the round one attacks, there’s a good chance SOME “safe” attack ends up NOT being safe after actual dice results then needing to withstand the opponent counter.
So there, let’s say you’re going with a somewhat lowball estimate of USSR survivors in a W Rus only open with fighters to Szechwan and Egypt. As I wrote, if Germany wants to take a shot at West Russia, if they go all in, what was it, they have 44%? It’s not bad odds. Conservative players won’t take it, but players like me often think it’s fun to just say “lol let’s do this!” then roll the dice.
There’s a lot of fun stuff I wrote about mathematical modeling of Axis and Allies results. One really fun thing I asserted is that I think most players think of outcomes as a single bell shaped curve. But actually attacker and defenders each generate a single bell shaped curve that don’t reinforce one another, they work against one another. So actually you get something that looks like a sine wave if you graph expected outcomes with the positive being attacker net survivors and negative being defender net survivors. But when you graph this switching defender survivors to a positive axis you get a two-peak curve.
Which means?
When I’m talking about a 44% result at West Russia, it’s NOT that “oh, there was a 56% chance Germany would lose but they ended up winning SO IT HAD TO BE CLOSE AND GERMANY PROBABLY JUST BARELY WON”. If you look at the mathematical model, if Germany DOES win, probably it’s NOT that close. If they win, probably it won’t be by just one or two units, probably it’ll be five or six or something awful. That’s what you get when you properly apply the two-peak curve mathematical modeling of Axis and Allies results.
Isn’t this really just fun stuff? I wrote a couple Steam guides for 1942 Online, go check them out. Probably I’ll never finish the third guide in the “basics” series, but eh. No collaborators makes for dull work.
