A defense of a defense

Anything about how you get those awesome core-dumps, or why you don't get them...
Post Reply
User avatar
Nanu Nanu
Core Dumper
Posts: 189
Joined: Wed Jul 27, 2011 3:20 am

A defense of a defense

Post by Nanu Nanu »

Image

Sine, just wanted to give my thoughts since I've seen you comment on the def I usually do a few times now, totally open to discussion. Yes it is slower to setup than a standard box, but the time difference is not significant, especially in a setting where you have a proper sweep setup (see some screenies below for a comparison between straight box def and mine). Usefulness is debatable, and of course I'm biased, but I feel it's an effective defense for me as a player who is not very mechanically skilled. It has 2 outward steps after each corner of the zone, and one inward bump in the middle. All of these turns are the same size, due to the low speed, which makes my life a whole lot easier on each rotation since I just have to make the same quick left/right, left/right, right/left/left/right. Something I hate is having an inward bump in my zone that is larger than standard, because I've got to then time my turn to be as close as possible to not leave too much gap (making the bump even bigger, causing rapid shrinking on each rotation there), or I turn too soon and die. Maybe this isn't an issue for other players, but it's something I struggle with. With a speed defense, when you begin to make bumps they can be irregular and large, not a fan. I maintain a low speed with all those starting turns which helps me maintain standard size turns each rotation and makes it easier for me to block cuts by braking and quickly stabbing my wall without dying despite being so close to it. The inward bump at the center of the zone, while slowing me down, is also to split up what I see as the main potential cut zone of my def, the long straight sections. If an attacker tries to lack attack on the first segment before the bump, I have the option of braking to mess with their timing, stabbing (which is hard against this attack and not my main choice), or racing them to my bump and stabbing if I am hiding a bit / just continuing my def and watch them slip out at the bump if I'm not hiding, since the second half of the bump is slightly smaller due to the speed reduction of the first half of the bump. Perhaps you could try to lack attack the second straight segment after the bump, but there's not much room to work with until that bump becomes larger later in the round, at which point I may have more setup. Most of the times when I get cut it's due to poor reaction speed or rubber mismanagement.

Main weaknesses to this def are: 1) a slow shrink as it's hard to expand without many long straight lines to speed on, which is primarily an issue later in the round. 2) Holing since the slow speed makes it harder to block in time. 3) Reliant on sweepers to setup properly in case the attacker can exploit the extra second it takes to setup.
On the first two points, it is my opinion that truly good attackers/players should always be able to slow shrink any defense and should always be able to hole without being thwarted by the defender alone. Also not saying this excuses me for doing a def that is particularly weak to these things, but they are tradeoffs I am willing to make to be as effective as I can at maintaining a def for a long period.

Regarding your comment "I think the bottle is obsolete. I rarely see it work and the time trade-off is a bad investment." There are varying levels of effectiveness depending on the bottle, but I definitely think it's necessary against good, aggressive centers. I don't think the time trade-off matters with a good team because the sweeps should be able to hold everyone off while the def sets up anyway, but of course you don't want to take forever on it. I do think you brought up a good point that an early split is a good center counter. It's a viable counter and in that case the def doesn't need to bottle, but doesn't allow for your center to go on the offense and try to kill the other center / try to center.

Here Andrei is doing a straight box def, on the other side I've just rounded the third corner, so he is ahead but not by much.
Image

Here's where I am when he's reached his tail
Image
Prema wrote:The second match starts, a new beginning,
Nanu and Prema, Sui and Ninja,
versus those same old hoes grinning.
syllabear
Shutout Match Winner
Posts: 1030
Joined: Fri Oct 13, 2006 1:37 pm
Location: UK/HK

Re: A defense of a defense

Post by syllabear »

I noticed (and briefly mentioned) your somewhat unique defence style. It is interesting to understand the rationale behind it.

I still believe step or some variant of step defence is probably the overall strongest defence, mostly due to its simplicity. I don't think a lot of time has been put into defence theory as it is not the most glamorous position. However, I think that relying on a particular type of defence leads to rigidity, and a good defender (and in reality, a good fortress and/or sumo player) should be ultimately flexible in their play. At the end of the day it all comes down to practise - the more you play, the better you'll get.

As for bottling - it really depends what your team is set up to do. If you centre is very aggressive and trying to centre-attack the enemy team, an early split will not be useful to this. In this case, a bottle is essential, to prevent the counter-centre. Obviously it is easier to set up without doing a bottle if the team decides to split early, but again, this comes back to what the overall goals of your team are.
The Halley's comet of Armagetron.
ps I'm not tokoyami
blondie
Core Dumper
Posts: 160
Joined: Sat Jun 21, 2014 2:57 pm

Re: A defense of a defense

Post by blondie »

There is logic to having "rounded" corners. I think two steps in the corner is too much, but one step is pretty clean. Basically the idea is that you want to have some speed breaks to conserve rubber and ultimately to conserve shrinking. Not all shrinking is equal, the corners you have a lot more you can shrink without exposing any zone to ganking, so putting your speed/rubber conservation there while keeping sides straight makes a lot of sense.

I don't love the 180 indents Nanu has at each side's midpoint. Those moves are good in a defense with square corners, but they ought to be closer to the corner, like between 3/5ths to 4/5ths of the way down a side. They are creating a bad shrink point by putting them closest to the zone.

Regardless, I don't good reason to set up your final structure on your first rotation. You can just do a square, which is faster, then put in your structure on the second rotation.
User avatar
sinewav
Graphic Artist
Posts: 6413
Joined: Wed Jan 23, 2008 3:37 am
Contact:

Re: A defense of a defense

Post by sinewav »

Nanu, thanks for screenshots of your defense compared to Andrei's. I can see the setup time is faster than I imagined. Reading your post, I don't think you give yourself enough credit. I've watched you play and I think you are a skilled defender! The rationale behind your defense is solid and I knew that's why you were doing it, but we can all agree on it's weaknesses. To extend Sylla's point about rigidity, focusing too much on the finer points of a trail could prevent a defender from predicting and responding to attacks. Of course, all the properties of a defense lie on a spectrum (fast/slow, passive/aggressive, etc.) and we can only do what's comfortable. I think you should have more confidence in your ability!

I think a lot about defense. So much that I wrote a book about it when CTF was my main game. My Fortress defense is extremely fluid and the direct result of me spending nearly all of my time looking somewhere other than my trail.
Word
Reverse Adjust Outside Corner Grinder
Posts: 4258
Joined: Wed Jan 07, 2009 6:13 pm

Re: A defense of a defense

Post by Word »

Imho the overall shape of the def isn't as important as how big you're going to make it from the start and how well the smooth sections where you speed up are balanced/"countered" with spaces to slow down: little steps or indents and where you decide to put "the four corners where you really have to turn" (meaning how you decide your def's proportions) to cover the whole circle of the zone.

There are three or four criteria for me that make a good defense:

1) you should be able to speed up and brake fast (or convert your defense during the course of the game if you need to, ideally without losing a lot of ground), but it's really ok already if you can go slow for a long time and keep the door shut,

2) you are at exactly the right spot at exactly the right time behind or in front of your trail end, or at least get there fast. I'm trying to convey here that this "right spot" constantly changes: You might need to expand or break out to box/core-dump an attacker at the right moment, or two guys are coming at you with lots of speed and are about to hole while your own team's attacker is about to get the other zone, so in this specific instance it's probably better to convert to a near perfect square and just go fast and stay so far in front of your trail end so that one side is always double-walled – if the situation allows you to do this,

3) there's no need to make it overly complicated so that you yourself can't navigate within your own walls once you get holed or boxed (clearly, I'm terrible at mazing, so I have to compensate that somehow),

4) depending on your internet connection you might prefer a simple def with little or no "inward indents" so an unexpected lag event doesn't necessarily kill you.

I'm somewhat agnostic about the use of bottling. Often you don't live to do a second rotation anyway when the sweepers are doing shit or dead trails cause a nice "air pocket" for the opponents to move in, but sometimes it seems to have a psychological effect on less experienced players and they don't try to center-attack when they easily could. I used to throw it in at random but I think finishing the first rotation and closing the whole thing fast is more important. I did it quite often when I knew that somebody in our team was afk or didn't know how to grind so I'd avoid the dead trail when closing the defense, but in a situation like that I'd also make random turns on the side that the opponents would likely reach last and then try to fix it one or two rotations later.

What all this comes down to is that I don't believe there's a perfect defense shape, you just keep your defense adaptable to specific situations all the time.
Last edited by Word on Thu Sep 10, 2020 9:17 pm, edited 2 times in total.
Olive
Match Winner
Posts: 501
Joined: Wed Mar 26, 2008 8:11 pm
Location: Amsterdam

Re: A defense of a defense

Post by Olive »

if you can post you can play

next ladle back to 16 teams? don't let your memes be dreams
Olive a.k.a ZeMu, MoonFlower & chicken.
Word
Reverse Adjust Outside Corner Grinder
Posts: 4258
Joined: Wed Jan 07, 2009 6:13 pm

Re: A defense of a defense

Post by Word »

Olive wrote: Thu Sep 10, 2020 12:52 pm if you can post you can play

next ladle back to 16 teams? don't let your memes be dreams
um nope, today i took a day off for a visit to the doctor. :) i have to get up at 4 am and get home like 16 pm, i'm already asleep when most of you are playing.
User avatar
Titanoboa
Reverse Outside Corner Grinder
Posts: 1795
Joined: Sun Feb 22, 2009 8:07 pm

Re: A defense of a defense

Post by Titanoboa »

Nanu's bottle is very good. There's no way to centre attack and survive the triple back at the end, which is possible with most quicker bottles.

The dents in the middle are significant against any attempts to cut, especially lackattacks. The setup is much more flexible than a step def in terms of expanding or recovering, as step defs tend to be left exposed for much longer should they ever need to break the pattern.

Nanu is virtually uncuttable.
User avatar
delinquent
Match Winner
Posts: 760
Joined: Sat Jul 07, 2012 3:07 am

Re: A defense of a defense

Post by delinquent »

Titanoboa wrote: Thu Sep 10, 2020 2:21 pm Nanu is virtually uncuttable.
Is that a challenge?
syllabear
Shutout Match Winner
Posts: 1030
Joined: Fri Oct 13, 2006 1:37 pm
Location: UK/HK

Re: A defense of a defense

Post by syllabear »

Titanoboa wrote: Thu Sep 10, 2020 2:21 pm Nanu's bottle is very good. There's no way to centre attack and survive the triple back at the end, which is possible with most quicker bottles.

The dents in the middle are significant against any attempts to cut, especially lackattacks. The setup is much more flexible than a step def in terms of expanding or recovering, as step defs tend to be left exposed for much longer should they ever need to break the pattern.

Nanu is virtually uncuttable.
Agree that it seems to be more resilient to specifically lackattacks and probably frustrating to play against for other speed type cuts, though I guess it depends what you're comparing to. I am not a good attacking player so I can't really comment. However it does seem to be much weaker against shrinking or hassle type attacks especially at the corners and indents, a weakness that step does not have. Yes you can recover more easily against these sorts of attacks, but at high level, the games do not last long enough for you to expand, and if they are, you've probably won that round
The Halley's comet of Armagetron.
ps I'm not tokoyami
User avatar
Nanu Nanu
Core Dumper
Posts: 189
Joined: Wed Jul 27, 2011 3:20 am

Re: A defense of a defense

Post by Nanu Nanu »

blondie wrote: Thu Sep 10, 2020 2:27 am I don't love the 180 indents Nanu has at each side's midpoint. Those moves are good in a defense with square corners, but they ought to be closer to the corner, like between 3/5ths to 4/5ths of the way down a side. They are creating a bad shrink point by putting them closest to the zone.
The main reasons I don't put the 180 farther down the side is 1) the segment before it is a longer straightaway and in my mind easier to cut and generates more speed, making the 180 larger than I'd like (unless I brake more of course, but would rather not use it much for this case). 2) I more quickly run into the third situation below, where I can no longer keep my 180 and end up a decent chunk behind my tail, which I've got to make up elsewhere if I'm trying to not hide too much.
I haven't noticed much that the slowly growing 180 in the middle is actually capitalized on until very late in the round, at which point most of my def is vulnerable, not just this part. Though I agree it's probably not optimal to have a growing shrink point close to the zone. I think I'll try having it closer to the end to avoid shrinking around the middle, and see how long I can make it work by braking. Just not sure what to do once I reach the third situation because I think those inward only steps are more easily cuttable.
Image
blondie wrote: Thu Sep 10, 2020 2:27 am Regardless, I don't good reason to set up your final structure on your first rotation. You can just do a square, which is faster, then put in your structure on the second rotation.
I have tried doing a square first to setup around the zone faster (tested in pickup last night after reading to refresh my memory!) but by main issue is that on the second rotation when I am setting up my structure there is a large gap, since my def is a good bit smaller than a standard box, leaving me vulnerable. I don't think it's too hard for the sweeps to keep attackers out while I setup a somewhat slower first rotation, but it's a bit much to ask that they continue to cover for me during a second rotation.
sinewav wrote: Thu Sep 10, 2020 4:24 am Reading your post, I don't think you give yourself enough credit. I've watched you play and I think you are a skilled defender!
Thanks! :) Also nice info on CTF def, wish I had read it when I was learning to defend there
Word wrote: Thu Sep 10, 2020 4:30 am...avoid the dead trail when closing the defense
This is another good reason for a bottle that I completely forgot about. I think compromising my second rotation due to dead tails is really rough and can have lasting impacts through the whole round. Also the dead trails leaving space is no problem with a more aggressive bottle!
Titanoboa wrote: Thu Sep 10, 2020 2:21 pm Nanu is virtually uncuttable.
Thank you for the high praise! As for how to beat my bottle, I've seen mr and fini turn into the offshoot to the right between 2 and 4, and do a small maze at the end, riding back up on the outside where my tail is disappearing (see screenie below). Currently I think the success of this depends on how long it takes 4 to grind, making the box at the end larger. Been working on some tweaks to deal with this center regardless of the grind, just haven't had anyone try for this lately :lol:
I wish I was uncuttable! There's plenty of proof of people simply being too fast for me to react or outplaying me when I try to go aggressive
Image
syllabear wrote: Thu Sep 10, 2020 3:13 pm ..it does seem to be much weaker against shrinking or hassle type attacks especially at the corners and indents, a weakness that step does not have. Yes you can recover more easily against these sorts of attacks, but at high level, the games do not last long enough for you to expand, and if they are, you've probably won that round
I don't think it shrinks much faster than a step def since I can usually keep my 180s the same size, but I haven't timed it so not entirely sure. There's also two routes I can take with it, one that I primarily do in pickup to stay close to my tail and look cuttable / have the option to attack, which I do by making more 180s at the 1/4 and 3/4 points of a side, causing me to for sure shrink faster than a step def on following rotations. The other route I do more in ladle where I keep the same pattern without extra bumps, slowly hiding more as the round progresses and not looking to go offensive / expand until later in the round, but it shrinks more slowly than the first option (imo). Mostly don't do this in pickup because it's probably boring to play against if I hide too much, and maybe they'll just hole me instead of playing with me :cry:. Also not sure what a hassle type attack is
Prema wrote:The second match starts, a new beginning,
Nanu and Prema, Sui and Ninja,
versus those same old hoes grinning.
User avatar
Titanoboa
Reverse Outside Corner Grinder
Posts: 1795
Joined: Sun Feb 22, 2009 8:07 pm

Re: A defense of a defense

Post by Titanoboa »

Nanu Nanu wrote: Thu Sep 10, 2020 5:16 pm There's plenty of proof ...

There's also two routes I can take with it, one that I primarily do in pickup ...

The other route I do more in ladle where I keep the same pattern without extra bumps, slowly hiding more as the round progresses ...
Yeah I was talking strictly about ladle, and post-110 especially. Even I've cut you a fair share in non-tourney fortress over the years to be fair :lol:

The only kind of cut (of your def in a ladle) I can picture is with a lot of chaos and many different walls around your def and some kind of surprise attack where you're caught off-guard. That never seems to happen anymore though. :P
Post Reply