0

Why do pro dota teams pick carry early like second pick in captains mode? Wouldn't picking cores last give more chances to counter?

For example this year in truesight LGD picked Tiny first in one game and lost and also picked Ursa second in another game they also lose. So everytime LGD picked core in first 2 pick they lose although 1 series is a small sample size.

3
  • You're always going to get countered though. Like, whatever role you pick first, opponents will be able to counter that hero.
    – Allure
    Oct 6, 2022 at 8:09
  • my point is that better to get a support countered than core
    – Lightsout
    Jan 11 at 5:18
  • Depends on how hard the counter is though, no? I'm not an expert on DotA, but I am relatively familiar with similar games, and in those games some heroes counter others quite harshly while others are only mild counters. A carry being mildly countered should still be better than a support being harshly countered.
    – Allure
    Jan 11 at 5:27

2 Answers 2

1

The support that goes with your carry best is easily countered

There may be a strategic advantage to picking carries early:

You're picking a team of 5 heroes. It may be the case that a different position has a tougher time when countered than the core position.

For example; let's say the following is true in isolation in patch X:

+--------------------------+---------+
|           Hero           | Winrate |
+--------------------------+---------+
| Antimage vs Shadow Demon | 40%     |
| Rubick vs Lion           | 70%     |
+--------------------------+---------+

Lion works well with Antimage in a lane. He supports the antimage until he reaches level 6 and then goes fight with the rest of the team. So the antimage team would like to pick Lion. If they do so first, then they give their opponents a better chance to pick Rubick, countering the Lion.

Versatility vs Linear strategy

If your carry/important hero happens to be quite versatile (fit in more compositions), but your support pick immediately gives away what degenerate combo/strategy you're doing ,enabling the other team to ban the other piece(s), then it can be a good idea to pick the more versatile part early.

For example, the Pudge (Played as #5) + Chen (played as #3/#4) combo is really good in this patch (something about fountains and hooks). Chen is a very versatile hero that goes in many compositions, but Pudge gives away what you're doing, so you would want to pick him last, even though he's supporting.

Or another example, let's say Tiny works well with Centaur. But Centaur is not very strong by itself this patch. Then you pick the Tiny first, even though Tiny is played as the #1 and Centaur is the #3 or #4 position. Picking the Centaur first gives away you want a Tiny, so the other team bans it ASAP, leaving you with a sub-par hero.

Or in another example, Nature's Prophet is quite good with a supportive Naga Siren (using Naga's ultimate ability twice with a Refresher Orb, then backdooring). But he can also work well with Broodmother (going for an early push). The two game plans are basically mutually exclusive: different item builds, different timings. It takes a while for the Naga to get to her Refresher Orb. Picking either of the two partners gives away the plan, so the opponent might pick Spectre to stop the backdoor plan, or Keeper of the Light against the push plan. But pick Nature's Prophet first, and they still don't know which of the two to pick, and it turns out Spectre is bad at defending an early push, and Keeper of the Light can't handle a late-game Prophet!

In reality, situations are more complicated than four heroes, there's a massive roster to consider. But this sort of illustrates the motivation a team captain can have to pick their heroes in a different order than 54321. It's all about minimizing information given to the opponents so they can't stop or harm your overall plan (as much).

A few choices are much more powerful, there is 'shortage of viable options'.

If the carry hero pool, that is: the pool of heroes with sufficient scaling to push through the enemy base, is very small, either due to the current patch causing long games, or due to imbalance, then there are only few 'viable' choices for the first pick.

If the carry is picked last or near-last, then it may be possible for the other team to lock you out of those few choices immediately.

An example:

In this patch, over 70% of games go over 45 minutes. Spectre, Medusa, Faceless Void, Naga Siren, Ember Spirit, Phantom Lancer, Alchemist1, and Lone Druid are the 8 'metagame' carries, and there are a couple more fringe viable heroes for the carry slot. As each team gets 2 bans before first pick, it's possible there are only 4 left. When the second set of bans swings around, they may very well all be gone. In fact, it's strategic to ban the carries here as you do need one in this farm-heavy patch.

1: Isn't technically 'carrying' but is spending 20 minutes buying scepters for the rest of his team.

2
  • I was thinking maybe the carry player has small hero pool. Seems to me like picking carry after second pick has slightly better winrate from watching the truesight this year
    – Lightsout
    Sep 30, 2022 at 21:48
  • @bakalolo That's a good observation, I added it to the answer.
    – aphid
    Oct 3, 2022 at 7:05
0

In the majority of tournaments, a meta develops where a handful of heroes are perceived as very difficult to deal with. Examples are Tiny and Magnus in TI10, Io (carry) and Tiny in TI9.

If you watch True Sight, you'll notice that the first pick and ban phase is often about denying those picks to the enemy team. Considerations like:

We don't want to deal with their rubick, magnus or tiny, so lets ban rubick and tiny and first pick magnus.

If one of the contested heroes is a carry and is available, that's what gets picked.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .