Lightning Axe is cheap removal you could consider that answers dangerous fatties and also enables Madness.
Noose Constrictor might be better than Wild Mongrel, depending on your playgroup. Do they play a lot of stuff where changing color matters (e.g. Terror effects, protection from green, Intimidate)? If not, having Reach is probably better.
You probably want the full set of 4 Reckless Wurm (or some Reckless Wurm with some Arrogant Wurm). 3-mana 4/4 Wurms and 4-mana 6/6 Wurms should be enough to win you the game without Dragons or Inferno Titans.
Garruk and Arlinn seem out of place. You don't need wolf tokens. Madness beatdown should get you there. If you want haste, go for 2 copies of Anger.
I agree Call of the Herd is a bit weak these days. It used to be such a powerhouse. 3 mana for a 3/3 isn't that great anymore. 4 mana for a 3/3 isn't great either. Getting both off one card is good value, but also slow and mana-hungry. It's better to pay 3 mana for a 4/4 trample or 4 mana for a 6/6 (Roar of the Wurm).
Stromkirk Occultist is a new card to consider, small, but it helps generate card advantage. You could also try Nimble Mongoose if your graveyard fills up fast (especially with Faithless Looting).
Izzet vs Dimir/Golgari. Game 3, after I punted the win in game 1 (my attack had him to below 4 and I had Inescapable Blaze in hand with 5 lands up, planning to finish him, but I clicked through priority on his kill spell to Goblin Electromancer without casting Blaze in response).
My deck had 2 Expansion // Explosions but I hadn't drawn either copy up to this point in games 1-3. I desperately needed one or the Blaze earlier to stabilize against his board, especially the 6/6 Molderhulk. Finally, with less than 20 cards in library, as the clock was ticking down and opponent was drawing through his deck, I went for the Hail Mary. Instead of casting Connive to steal his 2/4 Drake and try to race with Leapfrog (time-consuming), I used Concoct (Connive // Concoct) to revive Citywatch Sphinx just to chump Molderhulk. It was an elaborate way to Time Walk and Surveil 5, letting me dig 6 cards to sculpt my next draw step. It paid off. I found an Expansion // Explosion, drew it, and made him draw his deck for lethal at exactly 0:00!
I have a completely different read on your pool. To me this is an Izzet Pool where you can splash either black or white for value. Both decks look similar.
Charnel Troll is great, but both your black and green are shallow. Green's pretty bad in this format in general, and this pool is no exception.
You picked a Boros-seeded pack, but unfortunately your white is too weak to support the Boros aggro plan. Your white is Selesnya "go wide" stuff instead of the key mentor aggro stuff. Boros just looks bad here.
You could also try playing 5 color gates (running all your Gates, all your manafixing, Glaive, Guild Summit, and all your bombs). I think you don't have enough premium removal, card draw and bombs to really make up for running so much durdly fixing.
Why Izzet? Deep choices in red and blue. Removal and counters. Card draw. Murmuring Mystic just wins games.
Black splash gives you: Darkblade Agent, Notion Rain, Dead Weight, Severed Strands, Dimir Locket
White splash gives you: Justice Strike, Skyknight Legionnaire, 2 Garrison Sergeant, Boros Locket
There's some flexibility to swap a few spell slots depending on playstyle. Overall you're going for an Izzet tempo shell where you play cheap threats, disrupt their board with spells, and chip in some damage. Minotaurs + Sonic Assault and bounce will do a lot of work for you making it nearly impossible for your opponent to manage his defense.
Whisper Agent is really strong in this strategy. You can hold mana up for tricks all game (11 instant spells, 3 with Jump Start) and still EOT flash in a 3/2 (and fix your next draw!).
Discovery is a great card for card selection, even if you can't use Dispersal.
Radical Idea is pretty good with Electrostatic Field and the Minotaurs. It helps you cycle through your library and sculpt your hand while triggering the "instant and sorcery" stuff. I'd play 2 Ideas if it wasn't competing with so many other spells. It's also copiable with League Guildmage (essentially you can tap him to draw a card for 3 mana instead of 4!). All your cheap disruptive spells get better with League Guildmage.
As a controversial choice, I put Izzet Locket in the SB. It's a great card. I just like the other spells better with everything else your deck is trying to accomplish, and you don't really need the ramp mana except for Explosion.
3. Aggro Mill. This is where you start playing cards that only mill the opponent and don't have other value. The issue is that many of these cards are just worse than playing mono-red and going after life total. To win with mill you usually have to get rid of around 50 cards. Compared to the starting life total of 20 that means you need to hit 2.5 cards per point of life that a card of that cost would hit. Lightning bolt is one mana for three damage. At that ratio, a one mana mill card should burn 7-8 cards. Tome scour only removes five. At 2-3 mana burn is usually hitting for 4, for an equivalent 12 cards milled. Glimpse the unthinkable is only 10. This count makes Traumatize look really good, but 5 mana is a huge investment in a card that won't win you the game. You can't race other decks on those numbers. Burn is often the deck that sets the speed of the format, and you need some sort of game against it. I think the only mill card that really holds up here is Mind Funeral.
Really good point that needs to get emphasized more.
Mill isn't inherently bad. Wizards just nerfed the power level of mill cards compared to burn cards, so there aren't great options for going "aggro mill" (casting spells to mill your opponent's library as fast as possible until it's all gone).
They print Lava Spike and Lightning Bolt vs Tome Scour. To match Lava Spike in power level, Tome Scour would need to mill 7-8 cards instead of 5. To match Lightning Bolt in power level, it would need to read "Instant. Choose one - Target creature gets -0/-3 until end of turn; Put the top 7 cards of target player's library in his or her graveyard; Remove 3 loyalty counters from target planeswalker." Lightning Bolt just has so much utility. It can burn opponents, but it can also kill creatures and hurt planeswalkers, and it's an instant. Clearly Lightning Bolt is just so much better than any mill card ever printed, which is just unfair to the mill player.
Looking at the above math, Glimpse the Unthinkable is fair. If mill needs to be 2.5 times as much as burn and you can burn 4 for 2 mana (e.g. Boros Charm), then Glimpse is fair milling 10 = 4 * 2.5. The problem is Glimpse lacks flexibility. Boros Charm is an instant and also has 2 other modes, while Glimpse is a sorcery and can only mill. It's like the Lightning Bolt vs Tome Scour comparison above. Lightning Bolt and Boros Charm just do so many more things other than just attack the opponent.
These are the mill spells I would consider efficient for their mana costs: Archive Trap (0 mana -> mill 13) Hedron Crab (1 mana -> can mill 12+ over the course of the game) Memory Sluice (1 mana -> mill 8, if you have 2 cheap creatures) Glimpse the Unthinkable (2 mana -> mill 10) Trapmaker's Snare (2 mana -> mill 13, by getting Archive Trap) Mind Funeral (3 mana -> mill 8-14 ish, depending on how many lands they play) Sanity Grinding (3 mana -> mill 20+, but you have to build around it)
Boros was an absolute beating in Sealed with the seeded prerelease packs, but it might have been due to the skewed power level of the seeded packs. It might have also been because the DTB was Dimir (or BUr or UBg), so racing while they stumble to stabilize was the most viable strategy other than trying to grind out the mirror.
Since then it's gotten weaker but it still seems viable to me.
You need to build and pilot it like a proper aggro deck. Avoid 5-6 cmc unless it helps you win the game that turn. Running two 5 cmc "destroy target creature" is too slow for aggro. Demotion is where this deck wants to be. I think even casting Lightning Helix is a trap. I've gotten more wins casting it as the +2/+2 trick, with the option to Lightning Helix later if I run out of things to do with my mana.
As always with aggro, tempo and mana-efficiency are more important than card advantage. You can't win the lategame. The goal is to punish greedy decks for doing durdly things: keeping slow hands, losing tempo to multiple ETB lands, running too few bears, running awkward AABB cards, and tapping out on turn 3 for mana rocks or draw spells.
Attack as often as possible. Play multiple combat tricks to push through damage. You can't afford to go "oh he has a blocker, pass the turn". Have combat tricks. Bluff combat tricks. Turn things sideways. Blocking is for blue decks. Island.dec will outdraw you. Win fast or never.
Truefire Captain is probably a trap. RRWW is too inconsistent for aggro. As a Mentor curve-topper I like Barging Sergeant better, mainly for the easier mana cost and surprise factor. It's fragile, but it screws up their stabilization math.
Mentor's a great ability, but you need a good curve to Mentor. A bunch of 2/2 Mentor bears do nothing with each other. The best Boros decks I saw ran a mix of 1/x, 2/x, 3/x and buffs. They also sequenced their plays strategically, so they were consistently getting Mentor triggers each turn, often more than one per attack. If you don't get a few Mentors off early then the creatures fall behind in card quality, but if you get a few then its much harder to stop. Mentor is great if abused.
It's probably correct to run some 1 drops, especially Goblin Banneret and Healer's Hawk (Hunted Witness to a lesser extent). 1-drops can be mentored by anything. They also let you start mentoring on turn 3 instead of turn 4, a full turn of tempo earlier and a full extra trigger. 1/1s are bad topdecks, but Boros isn't trying to win topdeck wars.
Precombat buffs enable Mentor. Casting Take Heart or Sure Strike precombat can open you up for 2-for-1s and gives them more information for blocking, so it's normally a bad play, but it also enables another Mentor trigger. If you're applying enough pressure, they often need to tap out to play blockers and keep up. If they're tapped out but have blockers, I've gotten value doing the precombat pump to Mentor and get both attackers through safely (e.g. if you have 2/2 Mentor and 2/2 vs their 2/x).
I think this deck wants a lot of cards other decks don't want, which is advantageous in drafting. Stuff like 1/1s, Demotion, Maniacal Rage (useful either as Demotion or to buff a Mentor creature), Cosmotronic Wave... Even Maximize Velocity reads "R: Target creature gains +1/+1 and haste. Put a +1/+1 counter on target creature." I haven't resorted to playing it yet, but it doesn't seem that bad as a filler pump spell for what this deck is trying to accomplish.
I've noticed a pattern in several GRN Limited events I've played, both sealed and draft - I do well when I play three colors and poorly when I play 2. Is it just me?
That's been true for every Ravnica set, I think. The manafixing is very good and the power level of the gold cards is very good. In other words, there's high payoff for splashing and lower risk than usual.
It looks like I have this one in the bag (many ways to push through damage in hand).
Then he casts Inescapable Blaze at my face. Ouch. I go to 2. Cyclops gets +6/+0. He attacks with a 6/8 trample and 1/3 flyer. I can double block the Cyclops with Shieldmate and Truefire Captain, but would still take 1 trample for lethal. Only 1 mana open.
I cast Integrity//Intervention giving the Captain +2/+2. 6/5 + 2/2 now absorb all the damage and kill the Cyclops. I take no trample and survive at 1 life. Captain deals him 5. Then I swing back for lethal.
How does the autobuild for Sealed work? Does it just take the strongest 2 colors and then fill the deck with the top 23 cards in those colors? Does it do anything to try to control for curve or creature count?
Your program is good at picking colors, but makes questionable choices about deck construction. I think it's something that could be improved upon with some simple programming logic.
Here's a good example. The pool and deck the Autobuild gave me...
Only 12 creatures and only 1 2-drop. That's low for any deck, but even worse when you have 2 creatures that require sacrifices and 3 equipment. Also Double Act of Treason? Death Baron with 0 zombies? Gearsmith Guardian with 0 blue creatures?
I don't expect it to decide it's worth splashing green for the Elder Dragon and Goreclaw, or how that relates to the Dragon synergy with Spit Fire, the 4-power count, and the presence of Rupture Spire. Limiting it to 2 color decks is reasonable.
But there are some other things it could do better, like the number of creatures. 12 is too few in most cases (some archetypes in some formats are an exception). As a general rule, you could force a minimum number of creatures (e.g. 15) and have it pick creatures until it satisfies that number.
You could also try to control for creature curve. 1 2-drop is too few for most formats. The pool has more on-color 2-drops. They're filler, but filler is usually better than not having bodies at the right time. You could have it pick the best X 2-drops to ensure they're enough. You could also put caps on the number of 5s and 6+s to ensure the curve doesn't get too high.
You could control for the number of equipment and vehicles. 3 copies of Marauder's Axe is excessive, especially with 12 creatures and essentially no bears to equip. You could force a maximum of 2 equipment. That won't always be the right choice, but as a general rule of thumb I think it would lead to smoother auto-builds. In the rare exceptions where you want more equipment, users can tweak the deck.
I also think your algorithm has either overrated Act of Treason or needs to otherwise cap the number of spells that are card disadvantage. This pool has plenty of other options for the spell slots. Maybe if you forced the minimum 15 creatures Act of Treason naturally wouldn't make the cut and you'd be OK.
Here's what I built. It may not be the best build, especially since we don't know the format yet, but it seems better.
Accounting for all synergies is hard. But your algorithm could account for some basic synergies (especially in core set), like color-matters, artifacts matter, and tribal. You could build this into both the draftsim and the autobuild feature (if it builds the deck by iteratively selecting cards down the list).
If you're already in blue, Gearsmith Guardian goes up in value. If your colors are chosen and you're definitely not blue, Gearsmith Guardian drops to filler.
If you've chosen a tribal lord like Death Baron or a card like Spit Flame, most creatures of the tribe go up in value (the chaff like Infectious Horror are still bad).
This would involve each card having a certain number of attributes, not just the color and rating. I don't know how complex it would be to attach those features (e.g. color, type, tribe, equipment, mana cost), but I think that would help both the draftsim and the autobuilds by allowing the ratings to be modified by past choices.
1. Pay 1 mana to equip.
2. Sac goblin for R. Blade triggers and puts another goblin in hand.
3. Cast goblin for R.
4. Repeat 1-3
Cycle costs net 1 mana each time.
Ok, so if you can make each goblin creature card cost only 1 mana per body (1cc goblins or multiple bodies or cost reducers), then you can keep chaining it for each 1 mana you pay to re-equip. That means that if you have Skirk Prospector+Heirloom Blade (2 card combo + 4 mana to cast) and about 20 more mana, you can build a lethal Grapeshot.
Both your red and black are pretty weak. You're a bit short on legends to play Jaya's Inferno, but Squee helps.
I would just go to town with WB knights aggro. The 4 knights + Danitha are a dream team with Kwende. Between gang-blocking and Adamant Will, you represent the ability to stop most ground threats just with your bears while beating down in the air.
You're also forced to run a lot of historic filler, due to shortage of playables. Why not capitalize on the synergy. Black gives you Cabal Paladin for more profit and good 2 drops for Teshar to recur.
If you went with red, what cards would you play instead? Jaya, Squee, Fight with Fire, Run Amok, Rampaging Cyclops, and maybe a Radiating Lightning or Fervent Strike? Your curve looks better with black.
One of my favorite casual combos is tapping Pang Tong to give +0/+2 to Squire, transforming the meager Squire into a mighty Horned Turtle!! Imagine the huge tempo swing of attacking with a 2-mana Horned Turtle!
It's worth mentioning that the "Wheel" version combos off very quickly in multiplayer thanks to the insane mana generation from Waste Not when more than one opponent is discarding lands. As soon as you resolve one Wheel effect, not only will you draw a bunch of cards but there should be enough Waste Not mana to cast a Liliana's Caress and a Dark Deal or Ill-Gotten Gains immediately to continue the chain. You should have the gas to draw/discard a few times and hopefully kill them immediately with Liliana's Caress.
Using the Vintage B/R list, unpowered. This is far from optimal but should give a good idea of how wheels can be chained quickly for profit.
Instead of trying to go off immediately, you can build up a slower Legacy or Modern legacy multiplayer version with wheels. Build to a position where you wheel once or twice and then dump the Waste Not mana into a big spell like Torment of Hailfire or Death Cloud. After the Waste Not triggers resolve you should easily win that game.
Lightning Axe is cheap removal you could consider that answers dangerous fatties and also enables Madness.
Noose Constrictor might be better than Wild Mongrel, depending on your playgroup. Do they play a lot of stuff where changing color matters (e.g. Terror effects, protection from green, Intimidate)? If not, having Reach is probably better.
You probably want the full set of 4 Reckless Wurm (or some Reckless Wurm with some Arrogant Wurm). 3-mana 4/4 Wurms and 4-mana 6/6 Wurms should be enough to win you the game without Dragons or Inferno Titans.
Garruk and Arlinn seem out of place. You don't need wolf tokens. Madness beatdown should get you there. If you want haste, go for 2 copies of Anger.
I agree Call of the Herd is a bit weak these days. It used to be such a powerhouse. 3 mana for a 3/3 isn't that great anymore. 4 mana for a 3/3 isn't great either. Getting both off one card is good value, but also slow and mana-hungry. It's better to pay 3 mana for a 4/4 trample or 4 mana for a 6/6 (Roar of the Wurm).
Stromkirk Occultist is a new card to consider, small, but it helps generate card advantage. You could also try Nimble Mongoose if your graveyard fills up fast (especially with Faithless Looting).
Izzet vs Dimir/Golgari. Game 3, after I punted the win in game 1 (my attack had him to below 4 and I had Inescapable Blaze in hand with 5 lands up, planning to finish him, but I clicked through priority on his kill spell to Goblin Electromancer without casting Blaze in response).
My deck had 2 Expansion // Explosions but I hadn't drawn either copy up to this point in games 1-3. I desperately needed one or the Blaze earlier to stabilize against his board, especially the 6/6 Molderhulk. Finally, with less than 20 cards in library, as the clock was ticking down and opponent was drawing through his deck, I went for the Hail Mary. Instead of casting Connive to steal his 2/4 Drake and try to race with Leapfrog (time-consuming), I used Concoct (Connive // Concoct) to revive Citywatch Sphinx just to chump Molderhulk. It was an elaborate way to Time Walk and Surveil 5, letting me dig 6 cards to sculpt my next draw step. It paid off. I found an Expansion // Explosion, drew it, and made him draw his deck for lethal at exactly 0:00!
Someone must have been salty.
Charnel Troll is great, but both your black and green are shallow. Green's pretty bad in this format in general, and this pool is no exception.
You picked a Boros-seeded pack, but unfortunately your white is too weak to support the Boros aggro plan. Your white is Selesnya "go wide" stuff instead of the key mentor aggro stuff. Boros just looks bad here.
You could also try playing 5 color gates (running all your Gates, all your manafixing, Glaive, Guild Summit, and all your bombs). I think you don't have enough premium removal, card draw and bombs to really make up for running so much durdly fixing.
Why Izzet? Deep choices in red and blue. Removal and counters. Card draw. Murmuring Mystic just wins games.
Black splash gives you: Darkblade Agent, Notion Rain, Dead Weight, Severed Strands, Dimir Locket
White splash gives you: Justice Strike, Skyknight Legionnaire, 2 Garrison Sergeant, Boros Locket
1 Thoughtbound Phantasm
1 Nightveil Sprite
1 Goblin Cratermaker
1 League Guildmage
1 Vedalken Mesmerist
1 Electrostatic Field
2 Smelt-Ward Minotaur
3 Whisper Agent
1 Watcher in the Mist
//Spells: 12
1 Discovery // Dispersal
1 Radical Idea
1 Sure Strike
1 Disdainful Stroke
1 Unexplained Disappearance
2 Sonic Assault
1 Sinister Sabotage
2 Capture Sphere
1 Hypothesizzle
1 Expansion // Explosion
1 Izzet Guildgate
8 Island
7 Mountain
1 Selective Snare
2 Wishcoin Crab
1 Izzet Locket
1 Radical Idea
There's some flexibility to swap a few spell slots depending on playstyle. Overall you're going for an Izzet tempo shell where you play cheap threats, disrupt their board with spells, and chip in some damage. Minotaurs + Sonic Assault and bounce will do a lot of work for you making it nearly impossible for your opponent to manage his defense.
Whisper Agent is really strong in this strategy. You can hold mana up for tricks all game (11 instant spells, 3 with Jump Start) and still EOT flash in a 3/2 (and fix your next draw!).
Discovery is a great card for card selection, even if you can't use Dispersal.
Radical Idea is pretty good with Electrostatic Field and the Minotaurs. It helps you cycle through your library and sculpt your hand while triggering the "instant and sorcery" stuff. I'd play 2 Ideas if it wasn't competing with so many other spells. It's also copiable with League Guildmage (essentially you can tap him to draw a card for 3 mana instead of 4!). All your cheap disruptive spells get better with League Guildmage.
As a controversial choice, I put Izzet Locket in the SB. It's a great card. I just like the other spells better with everything else your deck is trying to accomplish, and you don't really need the ramp mana except for Explosion.
Goblin Electromancer (which also makes your spells cheaper)
Electrostatic Field
Young Pyromancer
Thermo-Alchemist
Thing in the Ice
I would avoid creatures that cost more than 2 mana.
To help sculpt your hand, you want stuff like Ponder and Preordain.
Izzet Guildmage and League Guildmage are creatures that also help you copy burn spells. Use Nivix Guildmage instead if you want to copy a high cost spell.
Really good point that needs to get emphasized more.
Mill isn't inherently bad. Wizards just nerfed the power level of mill cards compared to burn cards, so there aren't great options for going "aggro mill" (casting spells to mill your opponent's library as fast as possible until it's all gone).
They print Lava Spike and Lightning Bolt vs Tome Scour. To match Lava Spike in power level, Tome Scour would need to mill 7-8 cards instead of 5. To match Lightning Bolt in power level, it would need to read "Instant. Choose one - Target creature gets -0/-3 until end of turn; Put the top 7 cards of target player's library in his or her graveyard; Remove 3 loyalty counters from target planeswalker." Lightning Bolt just has so much utility. It can burn opponents, but it can also kill creatures and hurt planeswalkers, and it's an instant. Clearly Lightning Bolt is just so much better than any mill card ever printed, which is just unfair to the mill player.
Looking at the above math, Glimpse the Unthinkable is fair. If mill needs to be 2.5 times as much as burn and you can burn 4 for 2 mana (e.g. Boros Charm), then Glimpse is fair milling 10 = 4 * 2.5. The problem is Glimpse lacks flexibility. Boros Charm is an instant and also has 2 other modes, while Glimpse is a sorcery and can only mill. It's like the Lightning Bolt vs Tome Scour comparison above. Lightning Bolt and Boros Charm just do so many more things other than just attack the opponent.
These are the mill spells I would consider efficient for their mana costs:
Archive Trap (0 mana -> mill 13)
Hedron Crab (1 mana -> can mill 12+ over the course of the game)
Memory Sluice (1 mana -> mill 8, if you have 2 cheap creatures)
Glimpse the Unthinkable (2 mana -> mill 10)
Trapmaker's Snare (2 mana -> mill 13, by getting Archive Trap)
Mind Funeral (3 mana -> mill 8-14 ish, depending on how many lands they play)
Sanity Grinding (3 mana -> mill 20+, but you have to build around it)
Memory Sluice is pretty good with cheap reusable mill creatures like Hedron Crab and Grimoire Thief (and Conspire taps it!).
Sanity Grinding is good if you fill your deck up with a lot of blue mana symbols. Cards like Ghastlord of Fugue, Overbeing of Myth, Dominus of Fealty, Cryptic Command, Nightveil Specter, Counterspell, Tempest Djinn. It's hard to make that work in a pure mill deck, but you could just make a blue devotion deck with Sanity Grinding as a mill win condition.
Yeah, this in a nutshell is what Boros needs to do.
Since then it's gotten weaker but it still seems viable to me.
You need to build and pilot it like a proper aggro deck. Avoid 5-6 cmc unless it helps you win the game that turn. Running two 5 cmc "destroy target creature" is too slow for aggro. Demotion is where this deck wants to be. I think even casting Lightning Helix is a trap. I've gotten more wins casting it as the +2/+2 trick, with the option to Lightning Helix later if I run out of things to do with my mana.
As always with aggro, tempo and mana-efficiency are more important than card advantage. You can't win the lategame. The goal is to punish greedy decks for doing durdly things: keeping slow hands, losing tempo to multiple ETB lands, running too few bears, running awkward AABB cards, and tapping out on turn 3 for mana rocks or draw spells.
Attack as often as possible. Play multiple combat tricks to push through damage. You can't afford to go "oh he has a blocker, pass the turn". Have combat tricks. Bluff combat tricks. Turn things sideways. Blocking is for blue decks. Island.dec will outdraw you. Win fast or never.
Truefire Captain is probably a trap. RRWW is too inconsistent for aggro. As a Mentor curve-topper I like Barging Sergeant better, mainly for the easier mana cost and surprise factor. It's fragile, but it screws up their stabilization math.
Mentor's a great ability, but you need a good curve to Mentor. A bunch of 2/2 Mentor bears do nothing with each other. The best Boros decks I saw ran a mix of 1/x, 2/x, 3/x and buffs. They also sequenced their plays strategically, so they were consistently getting Mentor triggers each turn, often more than one per attack. If you don't get a few Mentors off early then the creatures fall behind in card quality, but if you get a few then its much harder to stop. Mentor is great if abused.
It's probably correct to run some 1 drops, especially Goblin Banneret and Healer's Hawk (Hunted Witness to a lesser extent). 1-drops can be mentored by anything. They also let you start mentoring on turn 3 instead of turn 4, a full turn of tempo earlier and a full extra trigger. 1/1s are bad topdecks, but Boros isn't trying to win topdeck wars.
Precombat buffs enable Mentor. Casting Take Heart or Sure Strike precombat can open you up for 2-for-1s and gives them more information for blocking, so it's normally a bad play, but it also enables another Mentor trigger. If you're applying enough pressure, they often need to tap out to play blockers and keep up. If they're tapped out but have blockers, I've gotten value doing the precombat pump to Mentor and get both attackers through safely (e.g. if you have 2/2 Mentor and 2/2 vs their 2/x).
I think this deck wants a lot of cards other decks don't want, which is advantageous in drafting. Stuff like 1/1s, Demotion, Maniacal Rage (useful either as Demotion or to buff a Mentor creature), Cosmotronic Wave... Even Maximize Velocity reads "R: Target creature gains +1/+1 and haste. Put a +1/+1 counter on target creature." I haven't resorted to playing it yet, but it doesn't seem that bad as a filler pump spell for what this deck is trying to accomplish.
That's been true for every Ravnica set, I think. The manafixing is very good and the power level of the gold cards is very good. In other words, there's high payoff for splashing and lower risk than usual.
In Sealed even 5 color is viable.
My life: 8
My board: Boros Challenger with +1 counter (tapped), Wojek Bodyguard (tapped), Vernadi Shieldmate (untapped) and Truefire Captain (untapped and summoning sick)
3 cards in hand. 1 land untapped.
His life: 12
His board: Muse Drake, Erratic Cyclops and 7 lands.
It looks like I have this one in the bag (many ways to push through damage in hand).
Then he casts Inescapable Blaze at my face. Ouch. I go to 2. Cyclops gets +6/+0. He attacks with a 6/8 trample and 1/3 flyer. I can double block the Cyclops with Shieldmate and Truefire Captain, but would still take 1 trample for lethal. Only 1 mana open.
I cast Integrity//Intervention giving the Captain +2/+2. 6/5 + 2/2 now absorb all the damage and kill the Cyclops. I take no trample and survive at 1 life. Captain deals him 5. Then I swing back for lethal.
Your program is good at picking colors, but makes questionable choices about deck construction. I think it's something that could be improved upon with some simple programming logic.
Here's a good example. The pool and deck the Autobuild gave me...
1 Doomed Dissenter
1 Death Baron
1 Plague Mare
1 Skymarch Bloodletter
1 Boggart Brute
1 Fell Specter
1 Brawl-Bash Ogre
1 Skeleton Archer
1 Gearsmith Guardian
1 Sparktongue Dragon
1 Epicure of Blood
1 Demon of Catastrophes
2 Shock
2 Explosive Apparatus
3 Marauder's Axe
1 Spit Flame
2 Act of Treason
1 Lich's Caress
//Lands: 17
9 Swamp
8 Mountain
1 Enigma Drake
1 Shield Mare
1 Luminous Bonds
1 Skilled Animator
1 Herald of Faith
1 Declare Dominance
1 One with the Machine
1 Gallant Cavalry
2 Pegasus Courser
2 Blanchwood Armor
1 Disperse
1 Essence Scatter
1 Aven Wind Mage
2 Aethershield Artificer
2 Giant Spider
2 Totally Lost
1 Divination
1 Omenspeaker
1 Titanic Growth
3 Child of Night
1 Anticipate
1 Salvager of Secrets
1 Sure Strike
1 Doublecast
1 Hostile Minotaur
1 Make a Stand
1 Daybreak Chaplain
1 Tormenting Voice
2 Fire Elemental
1 Viashino Pyromancer
1 Abnormal Endurance
2 Uncomfortable Chill
1 Frilled Sea Serpent
1 Gearsmith Prodigy
1 Inspired Charge
1 Vampire Neonate
2 Bone to Ash
2 Invoke the Divine
1 Rupture Spire
1 Crash Through
1 Infectious Horror
1 Wall of Mist
1 Root Snare
1 Walking Corpse
1 Goreclaw, Terror of Qal Sisma
1 Two-Headed Zombie
1 Vaevictis Asmadi, the Dire
1 Explosive Apparatus
1 Thud
1 Macabre Waltz
Curve:
1 - ssss
2 - csss
3 - ccccsss
4 - cccc
5 - cccs
Only 12 creatures and only 1 2-drop. That's low for any deck, but even worse when you have 2 creatures that require sacrifices and 3 equipment. Also Double Act of Treason? Death Baron with 0 zombies? Gearsmith Guardian with 0 blue creatures?
I don't expect it to decide it's worth splashing green for the Elder Dragon and Goreclaw, or how that relates to the Dragon synergy with Spit Fire, the 4-power count, and the presence of Rupture Spire. Limiting it to 2 color decks is reasonable.
But there are some other things it could do better, like the number of creatures. 12 is too few in most cases (some archetypes in some formats are an exception). As a general rule, you could force a minimum number of creatures (e.g. 15) and have it pick creatures until it satisfies that number.
You could also try to control for creature curve. 1 2-drop is too few for most formats. The pool has more on-color 2-drops. They're filler, but filler is usually better than not having bodies at the right time. You could have it pick the best X 2-drops to ensure they're enough. You could also put caps on the number of 5s and 6+s to ensure the curve doesn't get too high.
You could control for the number of equipment and vehicles. 3 copies of Marauder's Axe is excessive, especially with 12 creatures and essentially no bears to equip. You could force a maximum of 2 equipment. That won't always be the right choice, but as a general rule of thumb I think it would lead to smoother auto-builds. In the rare exceptions where you want more equipment, users can tweak the deck.
I also think your algorithm has either overrated Act of Treason or needs to otherwise cap the number of spells that are card disadvantage. This pool has plenty of other options for the spell slots. Maybe if you forced the minimum 15 creatures Act of Treason naturally wouldn't make the cut and you'd be OK.
Here's what I built. It may not be the best build, especially since we don't know the format yet, but it seems better.
1 Doomed Dissenter
2 Child of Night
1 Walking Corpse
1 Plague Mare
1 Skymarch Bloodletter
1 Boggart Brute
1 Death Baron
1 Goreclaw, Terror of Qal Sisma
1 Brawl-Bash Ogre
1 Skeleton Archer
1 Two-Headed Zombie
1 Demon of Catastrophes
1 Sparktongue Dragon
1 Fire Elemental
1 Vaevictis Asmadi, the Dire
2 Shock
1 Thud
1 Sure Strike
1 Macabre Waltz
1 Spit Flame
1 Lich's Caress
//Lands: 17
2 Forest
1 Rupture Spire
7 Swamp
7 Mountain
1 Explosive Apparatus
1 Marauder's Axe
1 Fell Specter
Curve:
1 - sss
2 - ccccss
3 - ccccs
4 - ccccc
5 - ccs
6 - c
Accounting for all synergies is hard. But your algorithm could account for some basic synergies (especially in core set), like color-matters, artifacts matter, and tribal. You could build this into both the draftsim and the autobuild feature (if it builds the deck by iteratively selecting cards down the list).
If you're already in blue, Gearsmith Guardian goes up in value. If your colors are chosen and you're definitely not blue, Gearsmith Guardian drops to filler.
If you've chosen a tribal lord like Death Baron or a card like Spit Flame, most creatures of the tribe go up in value (the chaff like Infectious Horror are still bad).
This would involve each card having a certain number of attributes, not just the color and rating. I don't know how complex it would be to attach those features (e.g. color, type, tribe, equipment, mana cost), but I think that would help both the draftsim and the autobuilds by allowing the ratings to be modified by past choices.
4 Skirk Prospector
4 Mogg War Marshal
1 Goblin Lackey
3 Goblin Vandal
4 Goblin Rabblemaster
4 Goblin Warchief
4 Goblin Ringleader
4 Siege-Gang Commander
4 Grapeshot
4 Ruby Medallion
3 Street Wraith
4 Gitaxian Probe
So what's the combo? Prospector + Blade?
1. Pay 1 mana to equip.
2. Sac goblin for R. Blade triggers and puts another goblin in hand.
3. Cast goblin for R.
4. Repeat 1-3
Cycle costs net 1 mana each time.
Ok, so if you can make each goblin creature card cost only 1 mana per body (1cc goblins or multiple bodies or cost reducers), then you can keep chaining it for each 1 mana you pay to re-equip. That means that if you have Skirk Prospector+Heirloom Blade (2 card combo + 4 mana to cast) and about 20 more mana, you can build a lethal Grapeshot.
I would just go to town with WB knights aggro. The 4 knights + Danitha are a dream team with Kwende. Between gang-blocking and Adamant Will, you represent the ability to stop most ground threats just with your bears while beating down in the air.
You're also forced to run a lot of historic filler, due to shortage of playables. Why not capitalize on the synergy. Black gives you Cabal Paladin for more profit and good 2 drops for Teshar to recur.
2 Knight of Grace
1 Mesa Unicorn
2 Knight of Malice
1 Benalish Marshal
1 Danitha Capashen, Paragon
4 Aesthir Glider
1 Cabal Paladin
1 Kwende, Pride of Femeref
1 Teshar, Ancestor's Apostle
1 Jhoira's Familiar
1 Serra Angel
1 Windgrace Acolyte
1 Demonic Vigor
1 Seal Away
1 Gideon's Reproach
2 Adamant Will
1 Blessed Light
//Lands: 17
10 Plains
7 Swamp
Curve:
1 - s
2 - cccccssss
3 - cccccc
4 - cccc
5 - ccs
If you went with red, what cards would you play instead? Jaya, Squee, Fight with Fire, Run Amok, Rampaging Cyclops, and maybe a Radiating Lightning or Fervent Strike? Your curve looks better with black.
Using the Vintage B/R list, unpowered. This is far from optimal but should give a good idea of how wheels can be chained quickly for profit.
4 Waste Not
4 Liliana's Caress
//Mana: 13
1 Sol Ring
1 Mana Crypt
1 Mana Vault
1 Lotus Petal
1 Lion's Eye Diamond
4 Chrome Mox
4 Dark Ritual
//Draw and Tutors: 8
1 Brainstorm
1 Ponder
1 Vampiric Tutor
1 Imperial Seal
1 Gamble
1 Mystical Tutor
1 Demonic Tutor
1 Tinker
1 Wheel of Fortune
1 Windfall
4 Dark Deal
3 Whispering Madness
3 Ill-Gotten Gains
1 Memory Jar
//Others: 2
1 Yawgmoth's Will
1 Elixir of Immortality
//Lands: 16
4 Darkslick Shores
4 Polluted Delta
4 Bloodstained Mire
2 Watery Grave
1 Blood Crypt
1 Steam Vents
Instead of trying to go off immediately, you can build up a slower Legacy or Modern legacy multiplayer version with wheels. Build to a position where you wheel once or twice and then dump the Waste Not mana into a big spell like Torment of Hailfire or Death Cloud. After the Waste Not triggers resolve you should easily win that game.