So this is the first in what could become a series where I look at which castles each house holds in a 7 Castle game.

It's a bit low tech - I took a pen and notepad and went through the most recent rated games whereby tyrell won BEFORE turn 10 - ie with 7 Castles.

Max turns 9 was the best filter I could come up with for 7 Castle only wins.

Secondly the sample size is rather small - 50 games but this already took a couple of hours so...

Lastly tides of battle games are included as well - again I didn't know how to filter them out so this is by no means exhaustive.

Without further ado here are the castles:

Highgarden and oldtown 50/50 - 100%
The reach 40/50 - 80%
Starfall 30/50 - 60%
Flints finger 22/50 - 44%
Yronwood 20/50 - 40%
Pyke 18/50 - 36%
Riverrun 17/50 - 34%
Sunspear 15/50 - 30%
Seagard 14/50 - 28%
Storms end 13/50 - 26%
Lannisport 12/50 - 24%
King's landing 11/50 - 22%
Harrenhal 10/50 - 20%
Crackclaw point & winterfel 6/50 - 12%
Moat cailin 5/50 - 10%
White Harbour 4/50 - 8%

This post is dedicated to ser hodor whose work has helped me become a much better player.

Some early observations...sniping Starfall is legit and highly recommended even if your dornish ambition goes no further.

Sunspear and storms end go hand in hand.

Flints finger is tyrells 5th best castle!

There is a big dropoff between taking Starfall only, yronwood as well and sunspear on top of it all. How deep should your dornish invasion go?

Conquering either greyjoy or martell seems to be the way forward! Don't try to take the eyrie.

1st and last games analysed were (ref) 261636 and 253440

---

Not today.

So some further thoughts:greyjoy is a better prospect than lannister. I suspect this is partly due to their lack of counters to text based abilities like loras Mace and qot. Also qot and Mace do counter balon to a certain extent.

If you do go after lannister it's best to do it through ironmans Bay otherwise you're fighting unreadable sea support - never a good idea.

Unreadable sea support is probably why baratheons castles feature so lowly in 7 Castle wins. The only house harder to conquer seems to be stark.

Whilst there is a lot of variance in which castles crop up the spread on the greyjoy Castles is a lot tighter than the dornish ones. Once you have Pyke it's a lot easier to take seagard, flints and riverrun it seems than to expand from Starfall to yronwood and sunspear.

Also making war on lannister does not seem to be sensible. While you have 3 close neighbours in lannister baratheon and martell it seems to make sense to pick either greyjoy or martell and go to town.

Also even when friendly with martell always be ready to snipe Starfall.

---

Not today.

Interesting to see, but I'm curious why you limited things to 7 castle games?  Are all wins not relevant, or does it just make things more manageable?

I am not surprised that Greyjoy is a more viable target than Baratheon/Lanni, as I have long been saying that their lack of counters to special cards makes them more vulnerable, and that fighting Bara is largely futile.  I am surprised to see Stark castles feature a bit more prominently than I expected.  I like nothing more than rolling right through Sunset and up to Winterfell (Stark is similarly lacking in counters to Loras or QOT), but I didn't think it was a particularly popular or viable path to attempt.

With regards to limiting it to 7 Castle wins I wanted a more pronounced victory. The idea is to think like a chess player navigating towards a favourable endgame.

While I agree all wins are relevant I wanted to see how tyrell can not just win but win big and derive some strategic signposts for tyrell to aim for rather than just 'I can loras x this turn'.

Also a castle win on tie breaks leaves less margin for error than a 7 Castle win imo. While I'd need to analyse 20 times more games to have a statistically significant sample I was looking for the biggest insight bang for my time buck.

---

Not today.

Very interesting, and as Hodor says, confirms that Grey / Martell are better targets than Bara / Lanni. Good to have numbers confirming that.

Thank you for this work !

---

Judge
LoI&F: Seasons 10 and 11 King
France WCup team
2021: Bronze
2022: Silver

Thanks guys. To a new player tyrell has 3 obvious houses to attack...looks like 2 of them are the wrong choice!

---

Not today.

I think from long ago that the Rose is the worst target and the worst threat for a Kraken - seems confirmed. A slight boost to Tyrell may actually be the strongest counter to the current Greyjoy dominance.

---

Judge
LoI&F: Seasons 10 and 11 King
France WCup team
2021: Bronze
2022: Silver

Considering the only two castles that never showed up in the 50 games analysed are the eyrie and dragonstone I would speculate that baratheons core 3 castles - namely crackclaw Kings landing and dragonstone show up very seldom except for martell.

With regards to buffing tyrell and nerfing greyjoy it would be interesting to see what a change in meta does to the results. I have to confess I've played many games as tyrell whereby its turn 3~4 and I'm not fighting anyone and genuinely at a loss to what to do next

---

Not today.

If it is turn 3-4, you do not fight anyone yet and can freely pick your target that actually is a great thing!

Who to pick is not straight-forward though and strongly depends on Westeros cards and board position.

---

“The only way to treat disease is to understand disease.“

It's counterintuitive, but more often than not the best choice for who to attack is whichever house is looking strongest at the moment.  If you do so, you will gain a (at least somewhat) more reliable ally in whoever they had been beating up on.  That weakened house will be more greatful for the relief, and also need a bit more time to reestablish themselves before your new dominance forces them to fight back.  Even if it is just a few moves worth, those can be crucial for fully wiping out the first enemy, securing your sea positions, etc.

Whereas when you join the dogpile on the bully's side, they are often strong enough to force you out as soon as you are done helping them, before you have a chance to establish a proper foothold in the territories of the eliminated player.

And you have also the sea factor : in the longer run, who controls the sea controls the land thanks to the support orders. A Tyrell attacking early Lannister on land makes no sense : it will in the long run benefit the Greyjoy.

---

Judge
LoI&F: Seasons 10 and 11 King
France WCup team
2021: Bronze
2022: Silver

Again the main conclusion for me seems to be that ironmans Bay is an absolute meal ticket if you can take it. The data almost suggests it is to tyrell what the narrow sea is to baratheon.

How many tyrell players start turn 1 saying to themselves 'if martell doesn't attack I'm working towards taking ironmans Bay as my plan A'?

Which house do you think I should do next? Lannister or martell?

---

Not today.

Baratheon

---

Judge
LoI&F: Seasons 10 and 11 King
France WCup team
2021: Bronze
2022: Silver

I think the comparison between Ironman and Narrow Sea is a bit off because Narrow Sea is literally a single march from Baratheon's starting sea territory, while Ironman is 3 away from Redwyne Staits.  But as WSS is definitely Tyrell's "home" sea area, and Loras can shorten those distances, it's not that far off.

The hard part is keeping 4-5 sea areas locked down consistently, as you need to continually defend Redwyne, WSS, and Sunset to have a chance to also hold onto Ironman/Golden Sound.  This is especially dicey if Martell is not thoroughly embroiled with Baratheon.  My way of thinking about it is not to worry about establishing super strong support networks, but instead continually have strong sea marches set out to respond to incursions from any direction. Just accept that you are going to have to continually battle for the seas, but those are battles you can win with Loras and Garlan helping to take a toll on enemy forces and reposition yourself to always be ready to counterattack against a strike on your flank by Martell or Stark.

Side note : Redwyne may be the only home sea area that you actually can leave empty. Of course, not having Oldtown accessing sea is slightly annoying, but if you enter IMB / GS / Sunset / WSS you don't want to have to split your 6 ships between 5 areas, 4 is way better. But it's better with a Martell knocking on Bara's door.

---

Judge
LoI&F: Seasons 10 and 11 King
France WCup team
2021: Bronze
2022: Silver

A truly insightful analysis, thanks PMeisterGeneral

As already mentioned by others, the results confirm that invading Greyjoy is a reliable way to reach a total of 7 castles when playing Tyrell. I witnessed this several times on thronemaster.net, but I never got the opportunity to do it myself unfortunately.

This strategy is not so intuitive though, especially as Lanni ships can cut your sea network. In my opinion it's better to meet some of these requirements before attacking Greyjoy:
- Greyjoy should struggle a bit against Stark or Lanni and preferably lose the sword too, so that his seas are weakly defended. A Web of Lies or a Storm of Swords can also be enough to weaken Greyjoy at sea.
- Martell should be friendly
- Lanni should be busy fighting on land
With a combination of 2 or 3 of the above, Tyrell gets a real opportunity against Greyjoy. As Ser Hodor stated several times on this forum, Greyjoy's combat cards get somewhat weaker during late game: lack of swords, no possibility to remove an order etc... That probably explains a lot about Tyrell success rate vs Greyjoy.

Which house do you think I should do next? Lannister or martell? I would also like to know the results for Baratheon, there are so many ways to go with this house! The route to 7 castles seems a bit more straightforward for Lannister and Martell. I guess Lannister goes after Greyjoy most of the time, and Martell wins are probably evenly split between Baratheon and Tyrell castles.