Distilled down to its fundamentals, Wars of the Roses is an area majority game. You try to control locations and personalities, which give you control points. These control points determine who scores victory points for each of the six regions on the board.
There are cities, towns, ports, nobles, bishops and ship captains that gradually come into play at the start of every round. Players simply take turns drafting them into service. They provide both income and control points. Money is important for raising troops for battle and for bribing and anti-bribing. Control points determine area majority and thus scoring. In the same round that a personality or location enters play, it can already be bribed away or attacked.
What's most interesting about the game is the secret planning phase, where you plan to attack enemy locations or defend your own locations, bribe enemy personalities to join your side or protect your own personalities from being bribed, and move your personalities around the board, which will impact where you can attack and the strength of your presence (control points).
Battles are simple. Every soldier kills one enemy and whoever has soldiers left wins. If two different players happen to attack the same location, the player who has a higher score attacks first and resolves the first battle against the defender. Whoever wins will have fewer troops left and will then have to face the second wave. So, like Power Grid and Inca Empire, there is a catch-up mechanism.
Bribes work in a similar way. You secretly spend money to protect your people from bribery. If someone else wants to try to bribe your personality, he would have to pay double of what you'd have to pay. Your anti-bribery incentive only protects against the first bribery attempt. If more than one opponent bribes the same personality, the last player to bribe gains the personality. Similar to attacks on locations, trailing players do bribing last.
Scoring is done at the end of every round. Presence in each region is compared and points are awarded. In the full four-player game, there is a parliament phase to elect the king, and players from the same side as the new king (York or Lancaster) gain bonus points. There are a few other ways of earning bonus points, e.g. having the most ships, controlling the most bishops. These are handled in a similar way as the largest army / longest road in The Settlers of Catan. This aspect creates another cause for competition.
So far I have played two games. The game starts slowly, since there aren't many locations and personalities at first. There isn't much to ponder over. You don't have much money to spend on bribery or battles anyway. Things get more interesting as the game progresses. There's more to think about when picking the newly available assets, in picking where to bribe/fight/defend, and in how to move your personalities to compete for control. The secret planning phase means there can be a lot of double-guessing. When I played I tried not to think too long. This is an aspect which can really drag if the players over-analyse. You'll never have enough money to protect all your assets or to make all the attacks that you want. So you need to be selective. I decided to mostly spend my money on offense, because it was just too hard to anticipate where attacks would come. Money spent on defense can easily be wasted when all the assets you protect don't get attacked/bribed. I spent on defense only when I knew I was counting on a particular asset to control a particular region that I wanted to win.
Maybe it was because I was the rules explainer, I did OK in round 1 (of 5), and then suddenly in round 2 things went very very bad. That was the result of being perceived as the biggest threat. One good thing is if you are very far behind the leading player, you can use a once-per-game French aid token, which gives you money depending on how far behind you are. I received $21 for this. That's how far behind I was. The money did help a lot. I even managed to bribe a very expensive noble, which I don't think anyone else expected. I gradually crawled back up, and later in the game it was Allen who fell behind and received a wagon of French francs.
Towards end game, we realised the importance of gaining the most points with the least effort possible, e.g. earning a second place score with minimal presence. We also prioritised, and even vacated some areas which we knew would be very difficult to win. The mobility of the nobles and ships was quite important. Also nobles tended to have very high control points. I won the game at 138pts. Han had 133pts, Allen 102pts. This was close.
Our second game was nowhere near close. Han focused on money-earning assets early (ships, towns, cities), and by mid game sprinted far ahead of both Allen and I. Allen picked up many mercenaries early, but delayed using them until mid game, when he launched many attacks simultaneously. However it might not have been a good idea, because by then he had lost most of his holdings except for some in the northen kampung (backwater) region, which scored very low. He fell behind in victory points and never managed to catch up. By the last round, Han was so far ahead that we knew it was hopeless to try to catch up. So I decided to instead focus my attacks on Allen, just to make sure he didn't overtake me. I spent all my money on bribing two of his very senior nobles, which he hadn't expected. By game end I was happy enough to retain 2nd place.
In that particular game session Han was really on fire, winning game after game and by big margins. Allen and I joked that he must be wearing red underwear (it is a superstition that red underwear brings good luck and negates your opponents' good luck). Final score: Han 177, me 106, Allen 82.
Although I generally don't fancy area majority games, I enjoyed Wars of the Roses well enough. The secret planning and double-guessing (of attacks, defenses, bribes and also very importantly movements) is the most unique part of the game. It's not all about lucky guesses. You do know roughly how much money your oppenents have and from looking at the board you can try to guess where they may want to compete. The choice between offense or defense is interesting. Offense is expensive, but defense can be a complete waste of money if you don't guess right. The overall package is pleasant and interesting.
I have thought about designing a game with a betrayal element. You manage a group of characters to do various tasks for you, but other players can bribe them away from you. You can spend resources to make them stay loyal to you. I never went far with the design. I think Wars of the Roses implemented this loyalty / betrayal aspect elegantly.
I think the game will be best with 4 players, because of the additional aspect of electing the king. There is a little team play aspect.