Skip to content

Splendor Strategy: Tactical Analysis (Gem Distribution)

A few posts ago I explored the two popular tactics in Splendor and which one is best in which situation. Was it building an engine and going for lower cards, or going straight for the big wins? The simple answer is it depends on the situation.

This time around we are going to explore the various card breakdowns of Splendor to see which one is worthwhile. Firstly, each set of cards (Sapphire, Emerald, Diamond, Ruby, and Jet) has the same kind of breakdown. Each resource is used 118 times across the three tiers of cards. In total, if you wanted to buy every card in the game (to become a gem magnate or whatever) you would need to spend 590 gems, for a total of 140 points. There are 28 points in each Gem set (so all Jets, for instance, are intrinsically worth 28 points). There are 35 cards worth no points, 15 cards worth one point, 15 cards worth two points, 10 cards worth three points, 10 cards worth four points, and 5 cards worth five points.

Those are the basic statistics, presented in the ever so clear way of putting them all in the same paragraph…yep…

Which Cards Are Best At Each Tier?

This is the big money question: Which cards are worth going after? The simple answer is it depends on the round as to which colour to go for; however the gem to point ratio is the same.

Below are the Tier One cards. Down the left hand side is the points value for the card. The X axis shows the number of coins needed to achieve the card. For these examples I am using Jet, simply because it was first on my list.

Splendor: Jets Tier 1

As it can be seen, there are a few immediate things that stand out. The first is there are two cards that only require 3 resources a piece. One of these needs just Emeralds. The other needs Emeralds and Rubies. Either one of these could be easy or difficult to get; however, starting the game they are only two moves away (Reserve Card, Collect two or three of a resource, Purchase).

That being said, only one card stands out as being truly amazing – there is a one point card that takes four of one resource for each Gem style in Tier One. For Jet it is Sapphire. Again, this is only three moves away (Collect 2, Collect 2, Purchase), however, it may be worthwhile reserving it first, just to make sure it remains yours. 

SIMILAR ARTICLE: Splendor Strategy: Noble Analysis and Noble Strategies

Splendor Jets Tier 2

Tier Two becomes somewhat more transparent depending on how Tier One went. If it is possible then it is worth collecting the one pointers, as at Tier Two they can really offer an easy choice on getting the three pointer. With the Tier One card, the three pointer only requires 5 resources, 4 if it is reserved first. These can be so worthwhile, and as can the 5 resource card (4 if it is reserved first), in this case needing diamonds. These two cards are equally difficult to get; however, the priority states that the suit you are collecting should go first. In other words, if you are collecting Jets then collect the Jet card first.

There are a few cards that aren’t overly worth it. At Tier Two it is not worth going for cards worth one point. This is because they cost twice as much as the one point cards in Tier One. If a one pointer comes out in Tier One, go for that instead.

Splendor Jets Tier 3

There are only two cards really worth going after in Tier Three. These are: the four pointer that requires only one resource, and the five that requires two. By this point you should have enough of one resource to offset the cost of the secondary resource on the five pointer. The four is easy enough to get.

The other four is not worthwhile. It requires one less of one type of resource than the other four; however, it requires 12 resources in total. Yes, a couple of these can be offset, but it just isn’t worth the gems.

The same with the three – 14 resources for 3 points? Definitely not worth it.

What is the Return on Investment for Each Splendor Card and Gem?

This then brings us onto the next section which is which cards deliver the biggest bang for their buck. For this I have had to configure the table slightly differently. We are removing the zero point cards, as they have their place in an Engine builder strategy, and would massively skew the table. Instead we’re looking at the best ROI per gem token in regards to points. For this I just copied the table below as it is an easier way to explain it.

Tier Gem Colour Points Diam. Sapp. Emer. Ruby Jet # Gems ROI
3 Jet 4 0 0 0 7 0 7 0.57
2 Jet 3 0 0 0 0 6 6 0.5
3 Jet 5 0 0 0 7 3 10 0.5
2 Jet 2 5 0 0 0 0 5 0.4
3 Jet 4 0 0 3 6 3 12 0.33
2 Jet 2 0 1 4 2 0 7 0.29
1 Jet 1 0 4 0 0 0 4 0.25
2 Jet 2 0 0 5 3 0 8 0.25
3 Jet 3 3 3 5 3 0 14 0.21
2 Jet 1 3 2 2 0 0 7 0.14
2 Jet 1 3 0 3 0 2 8 0.13

As it can be seen, there is a really good return on investment per gem of 0.57 for the Tier Three, four pointer. The best cards to go for, for ROI, are the following pictured below (in any suit/gem, not just Jet):


The best Splendor Cards to go for, in theory.

The worst card is the one point card, in Tier 2, that requires a huge 7 resources. This gives an ROI of a measly than 0.13. In other words, each resource is only worth 0.13 points. It is not a good investment.

What’s Left?

To be honest, I have now been staring at tables and graphs for way too long; however, there are probably a couple more things worth considering. Is it worth considering an aggressive tactic where you deliberately block another person from playing? I don’t think so, but it is worth looking at. Also, who can forget about the nobles? Which ones of them are worth looking at?

Both shall be explored in later blogs.

SIMILAR ARTICLE: Splendor Strategy: Noble Analysis and Noble Strategies

SIMILAR ARTICLE: Splendor Strategy: Blocking Other Players


  1. Note that the cards in Splendor are almost symmetric, but not completely. For example, there’s a 2-point card producing 1 blue which costs 5 blue, and likewise for green. There’s also a 2-point card producing 1 white which costs 5 red, and likewise for red/black and black/white. So red/black/white forms a 3-color cycle, while green and blue each form a 1-color (degenerate) cycle. It would be interesting to see bar charts for the 4 non-jet colors.

    If you’re building an engine, the 1-point 7-cost cards in tier 2 are probably easy to pick up (1-4 chips per card once your engine is running), and most of the time there will be one around. The cheaper 1-pointers in tier 1 are often crowded out by the 0-pointers.

    The 4-pointers in tier 3 which cost 6/3/3 are probably easy-ish to pick up by engines; likewise for the 5/3/3/3 for 3 points in tier 3.

    I’m not saying that engines are good; but once they’re up and running, they often mess with the cost/benefit/ROI math.

    Liked by 1 person

    • Engines in Splendor are definitely a good move, and I’ve never seen a game won without one, but you raise a good point about symmetry. Good point about using enginesfor the 6/3/3 and 5/3/3/3. I keep meaning to return to these articles, since they were the first I wrote for this blog, to elaborate on them a bit more.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: