If you missed last week’s post, click here to read Part 1 of the Bot Battles.
The Experiment
We are here to find out which bot played better in Bronze. So here is what I did. I started 2 new accounts on Splinterlands. I purchased an Archmage B token, and an Xbot Lite token and placed one on each account. I rented every Splinterlands card that was available in May of 2023. Since we are playing in bronze I did not rent any max-level promo card, because owning and renting these in Bronze is not cost-effective. I ran a minimum of 200 battles on each account, in both the Wild and Modern format. I collected data from Splintercards.com tracking the win percentage on each account. And I used the Splex.gg Omega Token to review the cards used and their win percentages. I hope the data is useful to you, and it improves your gameplay. Come back soon to see the results in Silver. That comes in Part 3!
Happy battling!
Dr. Zoom
The Costs
- $51.52 for the Archmage B Token
- $53.75 for the Xbot Lite Token
- $2.17 Complete Deck Rental Fees for 2 Days
- $0.08 Extra 50 Energies for Battle per Account
I elected to purchase the Archmage B token because it has lower delegation fees than the Archmage standard token (Archmage B is 25% compared to Archmage at 35%). I did not buy the Archmage A token (10% delegation fees) because it was more than double the cost of the Archmage B token. I realize that the savings might be worth it over time, but I wanted to start modestly. I purchases the Xbot Lite for the same reason; it was half the cost of the standard Xbot token. It only plays to Gold, but for a Bronze player, this is all they need. Regarding the card rentals I rented all cards except the 5 max level promo cards (Lux Vega, Waka Spiritblade, Prince Julian, Archmage Arius, or Mighty Dricken), because these are not cost-effective for purchase or rental when you are playing in Bronze. Renting these cards would have added about $1.50 a day in rental costs. That would add 50% to my costs, and in Bronze that is not realistic. I also only rented for 2 days to simulate the minimum a player would need to run the experiment. It also saved me time, because with over 300 cards, managing re-rentals is hard. I also purchased extra energy each day; at a cost of 1 DEC, this is was very reasonable to do.
Archmage Bot Settings
- Fees = 25% (Archmage collects 25% of your Focus and Battle Rewards)
- Frequency of Play = 4 (how many times per hour to play)
- MinCp = 5000 (What is the minimum Collection Power (CP) for the account to play?)
- Energy Target = 0 (Play match if above the chosen energy.)
- Harder Opponents = ON (Use this if you expect your opponents to have level capped teams for your league.)
- Aggression = OFF (How much weighting should the hand finder put into finding a team which matches the currently active focus.)
- No Banned Cards
- NO Focus rerolls
- Ban Starter Cards were NOT Banned (They could be used)
I ran the bot non-stop and added purchased energies when it was low. I ran the bot in Wild first. Once I had more than 200 battles, I switched the bot to Modern. I tried to mirror the setting as broadly as possible to match what an actual player may select to cover the most common scenarios.
Xbot Lite Bot Settings
- Experimental Mode = OFF (Includes the latest improvements to the team selection algorithm)
- Starter Penalty = 0 (Choose the best team regardless of the starter card penalty)
- Quest Priority = 0 (Choose the best team regardless of the quest)
- No Banned Cards
- No Focus Rerolls
I ran the bot non-stop and added purchased energies when it was low. I ran the bot in Wild first. Once I had more than 200 battles, I switched the bot to Modern. I tried to mirror the setting as broadly as possible to match what an actual player may select to cover the most common scenarios.
The Results
The results were fascinating, and I am excited to share what I learned. I was pleasantly surprised that both bots were pretty effective across leagues and their win percentages were solid. I feel that any player would do well with either of these bots. Battle help in the form of bots may not be for everyone, but for those looking to automate their play, increase their win percentage, or maximize rewards, Archmage and Xbot are helpful third-party services. These bots are effective, efficient, and excellent battle helpers.
Archmage Results
- Arch Modern Win % = 70.1%
- Arch Wild Win % = 85.6%
Archmage fared better in Wild than in Modern, posting about 15% more wins in Wild. Archmage has a much slower play style, posting a maximum of 4 battles an hour or at minimum 1 battle per hour. This may allow you to maximize your energy over the course of the day. It does spread out your battles throughout the day. This more paced approach is unique to Archmage. We ran 4 battles per hour recharging energy as we were able.
Xbot Results
- Xbot Modern Win % = 93.4%
- Xbot Wild Win % = 94.5%
Xbot’s winning percentage was extremely high, and this allowed my account to climb quickly through the ranks. The xbot plays battles quickly because it runs the battles back to back to back. This makes for very efficient battling and allows a player to move through the ranks rather quickly when they are winning. This can also have the opposite effect; should the bot lose multiple battles in a row, your account can lose rating points just as fast as you gained them. We ran 25-50 battles at time.
The Cards
In total, I rented every Splinterlands card that was available in May of 2023. Each deck or set includes all cards that were released, including reward and promo cards. Since we are playing in Bronze I did not rent any max-level promo card, because owning and renting these in Bronze is not cost-effective. This means I did not rent Lux Vega, Waka Spiritblade, Prince Julian, Archmage Arius, or Mighty Dricken.
These are the rental decks I used:
- Beta Deck for Bronze: https://peakmonsters.com/decks?deck=fc714794-ff47-4832-b97c-d41029cbe4f2
- Untamed Deck for Bronze: https://peakmonsters.com/decks?deck=3e112cec-3df3-4049-b4f0-da0d81e1d277
- Riftwatchers Deck for Bronze: https://peakmonsters.com/decks?deck=d928974a-b40e-412a-b781-b8b5ee0374c3
- Chaos Legion Deck for Bronze: https://peakmonsters.com/decks?deck=efdc4331-1b2f-4962-99d6-b9112686be8a
I also tracked the cards that the bots used. I own an Omega Token from Splex.gg which offers Prime access to my accounts. You can learn more at https://splex.gg/prime. I can also track my card usage for up to 30 days, and the win percentage per card. You can see the cards and win % below for each bot:
(Please note that the tabs at the bottom show the cards for each bot. So check both tabs.)https://docs.google.com/spreadsheets/d/1QclrMWxeEU9wq7sTGaNHv_sxaUrEFWjaHIxS3KYX9fM/edit?usp=sharing
Here is a rental link if you wanted to use all the cards used by Xbot:
These are the additional cards for Archmage:https://peakmonsters.com/decks?deck=939e5381-72dd-40d7-ac17-18438dbc89d4- https://peakmonsters.com/decks?deck=c7873924-865d-411c-99a5-bc1a3fc4936c
Conclusions
In the Bronze League xbot had a higher win percentage in both Modern and Wild formats, and is a strong option when considering automated play. Archmage was stronger in Wild with a very solid win percentage. Those playing the Modern format will find fewer wins with Archmage according to our data. Xbot used wider variety of cards (about 30 more), when compared to Archmage, in both formats. Both automatic playing systems would be helpful to players looking to play more matches and earn more rewards. While our experiment found xbot to have a higher win percentage, loyal users of either system will find success in Bronze with these bots. If you are looking to add a bot to your game play, these services are both good options. I would recommend that potential users check out the communities on discord and try them for yourself.
Thank you!
Thank you for being a part of the Splinterlands community and I hope all this data helps you to become a better player, have more fun, and earn more rewards. Best of luck to each of you. I will be running data in Silver next season. I hope you will follow us, as we put these bots to the test!
Happy Battling, Dr. Zoom
IGN: drzoom
Guild: Immortal Phoenix
Many thanks to Archmage, Splex, Splintercards, & Xbot for their great products. I also want to give shout-outs to splintercell-01, my guildmates, fellow players, and MAVs who encouraged me along the way!
Very well done experiment!
Thanks for the data !
You are very welcome!
Great stuff and thanks for sharing. Can't wait for when you try the higher leagues!
Just wondering what playing a full season would be like and how much difference there would be. Also, which period of the season did you do this experiement?
So, Archmage played most of the season, because to get to 250+ battles at 4 an hour you need lots of time. Archmage played from day 1 to day 12. Xbot played about the same, but since it runs in bursts I was able to get the data faster. I will say that I did run them last season (not included in this season data), and xbot was able to leaderboard in Wild in 2 days, but not modern, cause I ran out of time. Archmage was on its way, but did not have enough time to get there. I started a 0 rating btw for each. So, the data does span a range of time in the season. Also, I am running Silver next season, and maybe Champ too... you can see Mav Chat in discord for an idea someone pitched, so we will see if we can get Champ data too. And thanks for reading the article.
I started with Archmage-b on 4 accounts and migrated all 4 to xbot. After the fees of Archmage, I still have one Archmage-b saved. If one day they change the billing system or if the performance of xbot decreases, I can use it, but at the moment, I'm very satisfied with the service of xbot. Not only the bot itself, but also the card reports it offers, the other options like automatic card delegation (which I don't use), and even the automatic card rental even before the launch of peakmonsters deck system. Xbot is a great service with various features, in addition to its main feature
Very interesting! Those differences in win rates seem pretty significant to me although there are so many variables it's hard to tell from one run. How did rental cost fair against earnings?
I was talking to the Archmage folks and they noted that the Archmage bot has the most success in the higher leagues, which I thought was interesting. Regarding earnings, I was able to make leaderboard with the xbot, and Archmage was close. So, it you figure in all the costs, I was close to break even, or even net positive by a little with Daily and EOS chests. Really I think if you just use the best cards you can make a little with leaderboard earnings, otherwise you probably lose a little.
Yeah. Archamage focussed a lot on higher leagues to the detriment of lower ones, especially Gold Wild which suffered a lot after their V3 update. It seems that the bots in general favour modern cards, which make sense, since more people play them. It's likely to get worse when the modern ban comes in since the wild databases will record more and more modern cards being played. Most likely I'll be taking my wild cards out of ranked play and moving them onto lands.
I think that is a solid plan.
Congratulations @drzoom! You have completed the following achievement on the Hive blockchain And have been rewarded with New badge(s)
Your next target is to reach 100 upvotes.
You can view your badges on your board and compare yourself to others in the Ranking
If you no longer want to receive notifications, reply to this comment with the word
STOPCheck out our last posts:
We've done an in-depth comparison on XBOT vs Archmage performance here:
https://peakd.com/hive-13323/@splinterbank/special-offer-to-archmage-users-or-0percent-fees-for-30-days