--- title: Beat the Jobava 2 description: A full dive into the Jobava for Black image: https://image.lichess1.org/display?fmt=webp&h=550&op=thumbnail&path=SwYHLOG0qLky.webp&w=880&sig=eed23b7a88e62f0d736b5f0283a3144c088bf783 --- Now, that I have over a year of experience in playing against the Jobava with **3...a6-b5**, I think it's time to create a follow up to my [old study](https://lichess.org/study/H1VnSFnP/gdh15p6A) and [blog](https://lichess.org/@/CkickyCheck/blog/beating-the-jobava-london-and-other-nc3-lines/2SaSYfqJ). It's only **5 chapters**, so I'd reccomend looking through both of those before, to get a feel for the variation I'm going to show, if it's your first time seeing this. Back then I said White has **4 ways** of continuing their development: quick **Nf3–Ne5**, natural **Nf3–Bd3–O-O**, symmetrical **a3–b4**, or playing **a4**. That's kinda wrong. White has **2 more** unique ideas: **Bd3–Ne2** intending to meet **c5** with **c3** and delaying the development of the **g1** knight (waiting for us to play **e6**) in order to achieve **g4**. First let's take a look at the baseline — White simply castling short. ## Normal development https://lichess.org/study/ngw6LW8B/ARDxTVTJ#0 The move order is very important. We play **4...b5** instead of **4...e6**, so White can't get **g4** that easily. Playing **5...Bb7** would allow White to play **6.a4! b4 7.Na2!** attacking **b4** **7...e6 8.c3!** undermining the pawn. We want to meet that with **bxc3–c5**, but the bishop on **b7** is vulnerable (**Qb3**). **6...Bb7** is also important, as **6...c5** allows White to take and play **8.e4** (lines there get pretty [crazy](https://lichess.org/3xQPK7ra/black)). Having established the move order, let's look at the position. To an untrained eye it looks peaceful, boring even. Remember the thumbnail? The position where we are completely winning, despite for a moment being down **15 points** of material arrises from here! How's that possible? First let's look at what White can do. The obvious pawn break is **e4**. White can support that with **Re1** or **Qe2**. **a3** is also useful, to stop any **b4** intermezzi. Another plan for White is to try and use their develompent advantage to generate kingside play with **Ne5–Qf3**. Our next move regardless is **Nbd7**. It's the mose flexible, as we don't always want to castle, but the knight always goes to **d7**. ### Ne5–Qf3 *** https://lichess.org/study/ngw6LW8B/jgHy7Hvq#0 A very interesting idea of playing **11...g5** out of the blue. The queen on **h3** is rendered out of play. If White chooses **11.Qg3** instead, they run into a fork, losing the bishop pair, but the trouble doesn't end there, as the bishops can be immediately felt after **O-O–f5–Rf6–Nxe5–Rg6**. If white chooses to wait with **9.a3** instead — the same problem **10...g5!** with an attack brewing. And that's only an apperizer! The real magic after White goes for the **e4** plan supported by **Re1**! ### Re1–e4 *** https://lichess.org/study/ngw6LW8B/NnFX6YaM#0 Crazy right? **4 moves** and White can be lost from such a peaceful position. How likely is it to happen? Since discovering the idea I faced the *Jobava* once, and actually [got it in a game](https://www.chess.com/game/live/167776052318) already! Btw the position referred in the thumbnail arises after **9.a3 Nbd7 10.Re1** Now you might be wondering what can we do against **Qe2–e4**? White defends **f2**. Surely it's a boring draw right? ### Qe2–e4 *** > Honestly, this position deserves a study, or at least a chapter dedicated only to itself. If you aren't a GM with 90 minutes on the clock, don't go for this That's what I said over a year ago in my study against the *Jobava* about a line I'm about to show you. Even with **f2** defended White fails to avoid trouble! https://lichess.org/study/ngw6LW8B/AqIT23lE#0 Yeah... is there any value to analyzing the position so deep? Probably no. Was it fun? Yes! On some serious note, this sacrifice isn't as scary as it may seem. Remember, after **13...Qe7** White can't let us castle, so the queen has to stay on the **a2–g8** diagonal and since **Bd5** is a thing, the only move is **14.Qb3! Bd5 14.Bc4!** Then our idea is **Qf7**, but we want to gain some initiative on the queen. If White plays **17.Qb3** instead of **Kh1**, **17...Qf7** becomes much more dangerous, as the idea of **Re1–Re6** comes with only one threat — **Ng5**, which we can stop with **h6** initializing manual castling. Otherwise we get the queen trade. White has a much tighter rope to walk on, so I think the risk of **10...b4** is worth taking. The best part is White can avoid that by starting with **a3**. This is the tamest of all continuations, also the safest one for White. **11...d4** **12.Nb1 Nh5** is an interesting continuation, even with possible **f5** break, but it's very complicated, and with perfect play White get's very fast counterplay on the queenside with **13.Bd2 e5 14.b4–a4** From my research most people who play **e4** indent on pushing the pawn to **e5** if possible — therefore my reccomendation **11...Bb6**. We want to bait **12.e5** and strike with **12...d4**. Even if White navigates the complications correctly we still get an edge. Interesting position arrises after **12.exd5** aswell, where we recapture back with a pawn, getting an interesting IQP position, with **d4** on the table. Okay, with our baseline established, let's look at White's **Ne5** deeper. ## Ne5 – Madman https://lichess.org/study/ngw6LW8B/XTUbrYN8#0 Everytime this move is played White makes their intentions clear — an attempt at kingside attack. Whether it's with the queen, or **g4–g5** isn't yet revealed. I already covered what happens after **Bd3–Ne5–Qf3** with **O-O** included, is there a difference if White delays castling, or maybe goes long? ### Bd3–Qf3 *** https://lichess.org/study/ngw6LW8B/atjYFRnC#0 If White tries to get a *Sicilian* attack, they will run into more counterplay than they can handle, thanks to the open c-file. **9.Qf3** gollowed by **10.g4** is very interesting, as castling short quickly will land us in trouble, but there is a brilliant defensive resource **10...Bf8!** preemptively defending **g7**, so **11.g4** can be met by **11...Nxe5 12.Bxe5 Nd7!** Clearly with less space White can't comfortably launch an attack. An interesting way of dealing with that is going for a quick **g4**, hoping to get **g5** for free. ### g4 *** https://lichess.org/study/ngw6LW8B/Ep2MaoI3#0 **7...Nfd7** is a very strategical move, but funny thing is, even if White played something like **7.a3** or **7.Be2** it's still a great move, as the logic behind it stays the same — take on **e5**, if the d-pawn recaptures go for **Nd7–Bb7**, if the f4-bishop go for **Nc6** threatning **Nxe5** and when the bishop leaves **cxd4**. It wouldn't work against **7.Qf3** though, due to **8.Nxf7!!**, there you can play **7...Nbd7!** instead. ![image.png](https://image.lichess1.org/display?op=noop&path=pJm8xLky3fwj.png&sig=8545a665b9873a0b91afcacbdbb3b928d041e961)*After dxe5 you want the bishop on b7 due to the d5 pawn mobility, so the knight belongs on d7.* ![image.png](https://image.lichess1.org/display?op=noop&path=VX92_z3omfyo.png&sig=e5decd3b2705ba0bc80a7d45a96df0c32de5dff8)*With d5 blocked Bb7 isn't good, so blocking the diagonal doesn't matter. The bishop could get traded after a5–b4–Ba6 for example.* That should cover all fast **Ne5** tricks. Now let's look at a very unique approach — **a4**. ## a4 https://lichess.org/study/ngw6LW8B/AIo5ZLjK#0 White want's to quickly clarify the queenside structure in hopes to create a weakness, or freeze our play allowing for a free kingside initiative. ### Na2–c3 *** https://lichess.org/study/ngw6LW8B/MnFu6yuX#0 There is no other justification to playing **Na2** than following it up with **c3**. White can delay it, but will have to play it at some point. We want to meet that with **dxc3**. If White recaptures with the knight there are **2** weaknesses on **b4** and **b2** meaning the b-file control is crucial, but White's London Bishop defends the **b8** square, which means we HAVE to play **Bd6**. In case of **bxc3** we aren't in such a hurry. ### Ne2 *** https://lichess.org/study/ngw6LW8B/PIRSkFEA#0 It comes down to pushing **c4**, when we have a good answer to **b3**. If White were to play **8.Ng3 9.Bd3?** for some reason, we have **9...c4 10...a5!?** intenting to defend **c4** with **Ba6**, but we can't play **a5** before **c4** due to **Bb5+–Ne5**, so this only works when **c4** comes with a tempo. If White were to capture on **c5** all you have to watch out for is **Nd4–Bxb8–Nc6**. If you don't blunder that Black just stands better. **Nbd7–Re8/Qe7** is good, trying to play **e5**. If White goes for **Ne5** we as always take and retreat to **d7**. There isn't really much to the **a4** move. If White delays it with **5.Nf3**, **5.Bd3** or plays it even further down the line, the ideas remain the same. That's all about this unique approach. Now it's time to move onto the "critical" line with **a3–b4** ## a3-b4 – Strong opponent https://lichess.org/study/ngw6LW8B/frrjnYgd#0 It's quite rare. Out of **41 lichess + 26 chess.com** games In the Jobava I faced this **4 times**. If your opponent pulls up with this it likely means they are prepared. At least they think so, as with the right move-order you can likely trick them into a lost posotion. The best part is how often this trick works. Out of those **4 games**, I got it thrice: [2400 chess.com](https://www.chess.com/game/live/165243132270), [FM 2550 chess.com](https://www.chess.com/game/live/145396541112), [FM 2450 lichess](https://lichess.org/3NtXx2tR/black) and here's the [fourth game](https://lichess.org/uaXGwYnT/black). Out of curiosity I checked if GM Oleksander Bortnyk (a co-author of a Jobava course) has ever fallen for this trick, and to my surprise in the following position he blundered in **8 out of 9** games he played (on his chess.com account): ![image.png](https://image.lichess1.org/display?op=noop&path=SkLiO93JPz3q.png&sig=291ea34912c33e04150c0717640f6cc137373cc4)*The most natural 9.Bd3 is actually a blunder!* ![image.png](https://image.lichess1.org/display?op=noop&path=zXS0AyQoGuLF.png&sig=d161585504e48380157d1fe68979f2ac9f6c0c5b) How is that a mistake? First let's talk about the idea behind **a3–b4**. Similarly to **a4** White wants to clarify the queenside early, the difference being here we won't be able to play **c5**. Achieving **a5** is also difficult, because White has good pressure on **b5**. Which means all opening play will be in the centre and on the kingside. The downside of White's setup is the undefended **c3** knight, which would stand much better on **d2**. White aims to launch a very quick attack on the kingside with **Bd3–Nf3–Ne5–Qf3–Qh3–Bg5**, meaning Black can't afford mindless moves. Okay, now let's talk about how **9.Bd3** in the position above is a mistake! ### Nf3–Ne5–Bd3 *** https://lichess.org/study/ngw6LW8B/ZebzLoze#0 If White ever throws in **Bxe4**, we always take on **e4** first. The knight on **g4** will get trapped by **h5**. If it retreats to **f3** continue bulldozing with **e5** looking for a fork on **f2**, or **Nxg3–e4**. That means White should delay **Ne5** after castling. ### Nf3–Bd3–O-O *** https://lichess.org/study/ngw6LW8B/Lmai7vC8#0 There are more unique move-order tricks you can go for like **7.Nf3 Nbd7 8.Bd3 Ne4 9.Bxe4 dxe4 10.Nd2 f5 11.O-O? h5!** with g5 on the next move, but for the sake of simplicity I want to give White a chance to transpose into that lost variation. The position is quite dry, but it's possible to get a nice long-term advantage like the **e5** pawn. Similarly to White playing quick **a4**, here I don't have much more to say aswell. Next I want to tackle another idea often deployed by stong players, which is the move **g4**. ## g4 — True Jobava https://lichess.org/study/ngw6LW8B/2E6arAJy#0 Since we are delaying **e6**, White has to play smart to achieve it, waiting for us to play **e6** first. There are a couple of ways White can go about this — **5.a3**, **5.Bd3**, **5.h4** or even **5.Qf3**. We want to meet **g4** with **h6**, so if White wants to play **g5** White will have to play **h4** and defend the **h1** rook. The thing about this is it's just a good plan. The way we want to meet **g4** is **h6** slowing White down, then a quick **c5**. White will have to play both **g4** and **h4** to justify their unprincipled decision, so **5.h4** **h6** will transpose to a different **5th** **move**. Remember, we play **h6** so white has to play both **h4** and defend the rook in order to play **g5**, this will be important against **5.Qf3** Let's look at White's most common waiting move — **5.a3** ### a3–g4 *** https://lichess.org/study/ngw6LW8B/efxO4NNS#0 White can choose how they want to defend the rook and if they want to take on **c5**. We don't have a one for all cure, but I find very interesting, that against **Bg2** allowing **g5** is fine and even better for Black! The **8...Qa5** move is a novelty, so there isn't much analysis, but I really like how this move looks. Now let's look at what seems to be a much better way of playing **g4–g5–Bd3** — **5.Bd3** ### Bd3–g4 *** https://lichess.org/study/ngw6LW8B/JjYQCIss#0 I think a very practical choice **8...c4** should deal with this setup. If White takes on **c5** beforehand, we have a nice idea **9...h5!** seen in the last chapter aswell. On the topic of practical ideas White could try to achieve **g4–g5** with a very trichy **5.Qf3** ### Qf3–g4 *** https://lichess.org/study/ngw6LW8B/wJH4L7d0#0 Why don't we play **5...e6 6.g4 h6?** Because in this case it doesn't do anything. > Remember, we play **h6** so white has to play both **h4** and defend the rook in order to play **g5** Since the queen is already on **f3** defending the rook, after **7.h4** White already threatens **g5** and we aren't in time with our counterplay. That's why here I reccommend a slightly different approach with isntant **c5** "sacrificing" a pawn. That leaves us with White's last way of playing, IM Alex Banzea's ([upcoming Jobava London course author](https://alexbanzea.com)) reccommended **Bd3–Ne2–c3**. This is actually preferred by some strong engines according to my local analysis, but is extremaly rare in human play. The point is simple, White wants to solidify the centre before doing anyhing else. ## Bd3–Ne2 https://lichess.org/study/ngw6LW8B/Svi5hvJ6#0 The truth is I've never faced this, and it looks like a normal position. I don't see any value to analysing it deeper with an engine. Also this is a cool correspondence game: https://lichess.org/jp4ln5tP/black#0 That will conclude my deeper dive into the *Jobava London*. I'm really happy with the lines I found against White's natural development, taking on **c5** and playing for **e4**, where my games tended to liquidate. I also like positions abter **5.Nf3–Ne5**, especially my simple solution if White doesn't take on **c5** — **7...Nfd7**. Another setup I'm happy I've went through is White's quick **g4**, as that can be difficult to punish sometimes. I hope you found some value in this post, or maybe you got convinced to try the **3...a6–b5** line for yourself. In any case thank you for reading, leave a like if you enjoyed and see you next time! [Study link](https://lichess.org/study/ngw6LW8B).