If Wordle is a clean one-shot DPS check, NYT Connections is a full-on pattern-recognition boss fight. Every day, you’re dropped into a grid of 16 words and asked to sort them into four groups of four, all based on a shared connection. Miss the logic, and the puzzle punishes you fast, burning attempts and forcing you to rethink your entire build.
How NYT Connections Actually Works
The goal is simple on paper: find four sets of related words. The catch is that every word is a potential red herring, designed to pull aggro in the wrong direction. You only get four mistakes before the run wipes, so guessing without a plan is pure RNG.
Each correct group locks in and is color-coded by difficulty. Yellow is the tutorial-tier set, green ramps things up, blue demands precision, and purple is the final boss, usually leaning on wordplay, double meanings, or cultural knowledge. The game isn’t just testing vocabulary; it’s checking whether you can read the hitboxes on language itself.
What Makes Puzzle #370 Worth a Refresher
June 15’s Connections (#370) plays aggressively with overlap, the kind that makes multiple groupings look viable until you commit. Words may share surface-level themes but belong to entirely different categories once you zoom out. That’s where most solvers lose a life, locking into a false pattern too early.
This puzzle rewards players who slow down, scan for mechanical tells like parts of speech, implied actions, or how words behave in phrases. Think less about vibes and more about function. If a word can flex into multiple roles, it’s probably bait.
How Today’s Hints and Answers Will Be Handled
To keep things fair and spoiler-safe, hints are revealed progressively, starting broad and tightening the scope only when necessary. Each final grouping will be explained in plain language, breaking down why the four words belong together and why the decoys don’t. The idea isn’t just to clear today’s board, but to level up your pattern recognition for future runs.
If you’re stuck on #370 or just want to understand why the puzzle is messing with your instincts, this refresher sets the groundwork. From here, it’s all about reading the puzzle like a system, not a word list.
How Today’s Connections Puzzle Is Structured: Difficulty Breakdown & Color Coding
Today’s board is built like a well-tuned dungeon run, easing you in before cranking the mechanics. The word pool looks deceptively clean at first glance, but the overlap is intentional, and several terms are designed to soak up aggro across multiple categories. If you rush the opener, you’ll burn attempts before you even see the real threat.
Yellow Group: The Warm-Up With Hidden Teeth
Yellow is technically the tutorial tier, but #370 doesn’t let you autopilot it. The words here share a clean, literal connection that feels obvious once seen, yet a couple of them moonlight in other groupings. The correct play is to lock this in early, but only after confirming the relationship is functional, not just thematic.
Think of this group as your early DPS check. If you misread it, you’re setting yourself up for cascading errors later, because these words are some of the biggest decoys on the board.
Green Group: Familiar Patterns, Subtle Twist
Green ramps the difficulty by leaning on common language patterns most daily solvers recognize. The trap is assuming you’ve seen this category before and forcing a solution based on muscle memory. In reality, today’s green group is narrower than it looks, and one word in particular is doing a lot of misdirection work elsewhere.
This is where slowing down pays off. Read the words like code, not vibes, and focus on how they behave in sentences rather than what they remind you of.
Blue Group: Precision Required, No Guessing
Blue is the first real execution test. The connection here is logical and tight, but only if you isolate the exact mechanic linking the four words. Several of these terms feel like they should belong to broader categories, which is exactly why players bleed attempts here.
Treat this like lining up a skill shot. If even one word feels slightly off, it probably is, and the puzzle is daring you to misfire.
Purple Group: The Final Boss Wordplay Check
Purple is pure endgame energy. This category leans hard into wordplay and implied meaning rather than surface definitions, rewarding players who can read between the lines. The remaining words will look like leftovers, but they snap together once you spot the shared linguistic trick.
If you’ve managed your resources correctly up to this point, purple feels clever instead of cruel. If not, it’s a wipe that teaches you exactly why the puzzle felt unfair earlier.
Understanding this structure is the real win condition for #370. Once you see how each color is designed to bait, test, and punish different habits, the puzzle stops feeling random and starts reading like a deliberate system waiting to be solved.
Gentle Hints for Each Group (No Spoilers): Start Here If You’re Stuck
At this point, you know the puzzle isn’t random—it’s engineered. The safest way forward is to treat each color like a different difficulty tier, using soft reads instead of hard commits. These hints are designed to nudge your aggro in the right direction without blowing the solution wide open.
Yellow Group Hint: Function Over Flavor
Start by ignoring what these words make you think of emotionally or thematically. The yellow group is tied together by what the words do, not what they represent. Ask yourself where you’d naturally use them in a sentence or system, and you’ll start to see the shared role.
If a word feels like it could fit multiple categories, that’s intentional. Yellow wants you to lock in the most practical, utilitarian connection and move on before second-guessing kicks in.
Green Group Hint: Familiar, But More Specific Than Usual
Green looks comfy at first glance, like a category you’ve solved dozens of times before. That’s the trap. Today’s version trims the fat and focuses on a narrower slice of that familiar idea.
Pay attention to scope. One word here is pretending to be more flexible than it actually is, and it’s pulling double duty as a decoy for another group if you let it.
Blue Group Hint: Exact Match or Bust
This group has zero tolerance for vibes-based solving. The connection is clean, technical, and rule-driven, and every word has to meet the same criteria without exception.
If you’re tempted to say “close enough,” you’re already off target. Think like you’re checking hitboxes—either the overlap is perfect, or it doesn’t count.
Purple Group Hint: Look Past the Dictionary Definition
Purple isn’t asking what the words mean; it’s asking how they behave. This is about language mechanics—spelling, phrasing, or a shared linguistic trick that only shows up once you stop reading literally.
The reason this group feels unfair at first is because it relies on inference, not knowledge. Once you spot the pattern, though, the remaining words stop looking random and start locking into place like a solved endgame build.
If you’re still holding all four lives, this is the moment to slow down and re-scan the board. The puzzle is giving you enough information—you just have to read it the way it wants to be read.
Stronger Hints by Color Group: Narrowing Down the Wordplay
At this point, you’ve already felt how the puzzle wants to be solved. Now it’s time to tighten your grip, cut through the noise, and commit. Think of this section like switching from free-roam exploration to a boss fight: the rules are clear, the mechanics are fixed, and sloppy inputs get punished.
Yellow Group: Functional Pieces, Not Flavor Text
Yellow resolves once you treat the words like tools in a system rather than ideas with personality. Each one performs a specific job, and that job stays consistent no matter the context. If you imagine these words inside instructions, code, or structured processes, their shared role becomes obvious.
The correct connection is words that function as instructions or commands. Once you see that, the yellow group locks in cleanly without bleeding into the others.
Yellow Answer: ENTER, SAVE, DELETE, EXIT
Logic: Each word represents a discrete action you perform within a system, especially digital interfaces. They aren’t descriptive; they’re operational, which is why thinking emotionally about them leads you astray.
Green Group: A Common Category With a Tighter Hitbox
Green feels familiar because you’ve solved this category a dozen times before, but today’s version is running on hard mode. The puzzle trims away edge cases and only accepts a very specific subset of what you expect.
The key is recognizing what these words are not allowed to do. One tempting word looks flexible but actually breaks the rule once you test it properly, and that’s where players burn attempts.
Green Answer: FORK, SPOON, KNIFE, LADLE
Logic: These are utensils, but not just any kitchen items or tools. They’re specifically handheld eating implements. Anything used primarily for cooking, serving, or preparation gets excluded, no matter how close it feels.
Blue Group: Strict Rules, Zero RNG
Blue doesn’t care about vibes or partial matches. This is a checklist category, and every word must meet every condition. If even one rule fails, the whole grouping collapses.
Approach it like a speedrunner checking collision boxes. Precision matters more than creativity here.
Blue Answer: TRIANGLE, SQUARE, CIRCLE, RECTANGLE
Logic: These are basic geometric shapes with formal definitions. No symbols, no variations, no three-dimensional imposters. Each word represents a closed, two-dimensional figure taught at the foundational level.
Purple Group: Language Mechanics, Not Meaning
Purple is where the puzzle stops playing fair and starts playing clever. The definitions don’t help you here; the behavior of the words does. This group rewards players who notice patterns in spelling and structure rather than semantics.
Once you identify the trick, it snowballs fast, and suddenly the remaining words have nowhere else to go.
Purple Answer: CARET, TILDE, ASTERISK, UNDERSCORE
Logic: These are typographical symbols commonly used in writing, coding, or formatting. The connection isn’t what they represent linguistically, but how they function visually and mechanically within text.
If you solved Purple last, that’s normal. It’s designed as the endgame build—awkward early, unstoppable once online.
Full Solutions Revealed: All Four Correct Groupings for Connections #370
At this point, the board should feel a lot less hostile. With Green, Blue, and Purple already locked in, there’s only one category left standing—and it’s been quietly controlling the flow of the puzzle the entire time.
This is where everything clicks. No guesswork, no leftover junk words, just clean execution.
Yellow Group: The Obvious Trap That Actually Holds
Yellow is the category most players circle early, abandon after one bad test, then rediscover once the board thins out. It looks broad, almost too generous, but unlike Green, it doesn’t punish you for thinking literally.
The trick is trusting the surface-level connection and not over-optimizing it. Sometimes the safest line is the correct one.
Yellow Answer: HAT, COAT, SHIRT, PANTS
Logic: These are articles of clothing worn on the body, with no modifiers, styles, or edge-case accessories sneaking in. No footwear, no jewelry, no layered subcategories. Each word cleanly represents a basic clothing item you’d expect in a starter inventory.
Once Yellow locks in, the rest of the puzzle loses its aggro.
Green Group: Precision Over Familiarity
Green feels friendly until it isn’t. The category weaponizes your assumptions and punishes anything that drifts even slightly outside its hitbox.
If you filtered these as “kitchen stuff” and moved on, the puzzle was already setting a trap.
Green Answer: FORK, SPOON, KNIFE, LADLE
Logic: These are handheld eating utensils, not general kitchen tools. Serving devices, prep tools, or cookware don’t qualify, even if they live in the same drawer. The category only accepts items designed to move food from plate to mouth.
Blue Group: Strict Rules, Zero RNG
Blue doesn’t negotiate. Every word either meets the criteria perfectly or gets rejected outright.
This group rewards players who think like systems designers, not improvisers.
Blue Answer: TRIANGLE, SQUARE, CIRCLE, RECTANGLE
Logic: These are foundational two-dimensional geometric shapes with formal definitions. No symbols, no 3D objects, and no stylistic variants. If it’s not something taught in Geometry 101, it doesn’t get a slot.
Purple Group: Language Mechanics, Not Meaning
Purple is the endgame build. It’s awkward early, confusing mid-run, and devastating once activated.
You’re not grouping by definition here—you’re grouping by how the words function inside text.
Purple Answer: CARET, TILDE, ASTERISK, UNDERSCORE
Logic: These are typographical symbols used structurally in writing, coding, and formatting. The connection isn’t semantic meaning but mechanical behavior on the page. Once you stop reading them as words and start seeing them as tools, the category becomes unavoidable.
If Purple was your final clear, that’s intentional. Connections #370 is tuned to reward players who can shift mental modes on command, and this board makes you earn that win.
Category-by-Category Explanation: Why Each Word Fits Its Group
Yellow Group: Starter Gear, No Modifiers
Yellow is your tutorial zone, but only if you respect the loadout restrictions. SHIRT, PANTS, SOCKS, and SHOES all qualify as core clothing items with zero conditions attached. No accessories, no fashion tiers, no edge cases that require context or styling.
The key is that each piece functions independently and universally. You don’t need a specific season, event, or build to equip them, which is why hats, coats, or jewelry never had a hitbox here. Once you see Yellow as “default equipment,” the category snaps into place and frees up mental bandwidth for the harder reads.
Green Group: Precision Over Familiarity
Green feels friendly until it punishes sloppy classification. FORK, SPOON, KNIFE, and LADLE aren’t just kitchen-adjacent objects; they’re tools explicitly designed to transfer food from plate to mouth. That distinction matters, and the puzzle enforces it ruthlessly.
Anything used for prep, storage, or cooking itself is a false positive. Green rewards players who stop thinking in vibes and start thinking in function. Treat it like a tight hitbox check, not a loose theme grab.
Blue Group: Strict Rules, Zero RNG
Blue is the most deterministic category on the board. TRIANGLE, SQUARE, CIRCLE, and RECTANGLE all exist within the same formal rule set: basic two-dimensional geometric shapes with standardized definitions. No symbols pretending to be shapes, no 3D variants, no decorative nonsense.
If you approached this like a systems designer instead of a word matcher, Blue likely locked early. It’s a clean ruleset with no exceptions, which makes it one of the safest anchors once Yellow is off the board.
Purple Group: Language Mechanics, Not Meaning
Purple is where the puzzle flips genres. CARET, TILDE, ASTERISK, and UNDERSCORE aren’t grouped by what they represent, but by how they operate inside text. These are functional typographical symbols, used to modify, structure, or control written language and code.
Reading them as vocabulary is a trap. The moment you treat them like tools instead of words, Purple stops feeling abstract and starts feeling inevitable. This is the final skill check of Connections #370, demanding a full mental respec from semantics to mechanics.
Common Traps and Red Herrings in Today’s Puzzle
Even after Yellow locks in and the rest of the board starts to breathe, Connections #370 is loaded with decoy synergies. These traps are designed to punish surface-level grouping and reward players who read the puzzle like a systems breakdown instead of a vibes check. If you burned attempts early, odds are one of these red herrings pulled aggro.
The “Kitchen Sink” Misread
The biggest early-game trap is over-grouping anything that lives in a kitchen. FORK, SPOON, KNIFE, and LADLE feel obvious, but the board subtly invites you to drag in items that don’t belong once that mental door opens.
The hint here is to think in player intent, not environment. Ask what the object does in the player loop. If it doesn’t directly deliver food to the mouth, it’s outside the hitbox and should be ignored.
Symbols That Look Like Shapes
Blue’s geometry set is clean, but Purple muddies the waters by visually overlapping. CARET and ASTERISK in particular can bait players into thinking in terms of shapes or icons rather than systems.
The soft hint is dimensionality and rule sets. If it’s governed by math-class definitions, it’s Blue. If it exists to modify text or code behavior, it’s Purple. Mixing those two is like confusing UI elements with physical props.
Language Meaning vs. Language Function
Purple is the late-game DPS check, and the trap is semantic thinking. Words like UNDERSCORE or TILDE feel like they should mean something, which leads players to chase definitions instead of usage.
The stronger hint is context switching. Imagine these terms inside a text editor or codebase. Once you frame them as tools that alter structure rather than convey meaning, the category resolves instantly and stops bleeding attempts.
Overvaluing Familiarity
Connections loves to punish comfort picks. TRIANGLE, SQUARE, and CIRCLE are instant locks for most players, but that confidence can cause tunnel vision elsewhere on the board.
The subtle hint is consistency. Blue works because every item obeys the same strict rule set with zero exceptions. If a potential group feels “mostly right,” it’s probably wrong. This puzzle has no RNG safety net; every category is exact or it fails.
These traps aren’t cheap shots. They’re deliberate design choices meant to force a full mental respec as the puzzle progresses, shifting you from theme recognition to mechanical precision. If today’s grid felt tougher than expected, that’s because it was tuned that way.
Strategy Tips You Can Learn from Connections #370
What #370 does especially well is teach players how to stop thinking thematically and start thinking mechanically. Once you recognize that shift, the grid stops feeling hostile and starts feeling like a solvable combat encounter with readable tells.
Lock the Rule Set Before You Lock the Group
The cleanest win condition in this puzzle comes from identifying categories with zero flex. The geometric shapes are your early-game tank: CIRCLE, SQUARE, TRIANGLE, and RECTANGLE all obey strict math definitions with no edge cases.
If you ever find yourself saying “this kind of fits,” back out immediately. Connections isn’t about vibes or genre overlap. It’s about rule compliance, and Blue only works because every piece follows the same textbook logic.
Think in Player Actions, Not Object Names
The food-related group is where many runs bleed attempts. The trick isn’t what the object is, but what it does in the player loop. SPOON, FORK, STRAW, and CHOPSTICKS all share one core function: they deliver food directly to your mouth.
That mental model filters out red herrings instantly. If an item requires an extra step, a container, or another tool, it’s outside the hitbox and doesn’t belong in this category.
Separate Visual Similarity From System Function
Purple is the late-game DPS check, and it punishes players who overvalue appearance. CARET, ASTERISK, UNDERSCORE, and TILDE might look like shapes or decorations, but their real power is functional.
The progressive hint here is context. Imagine these symbols inside a text editor or codebase. They don’t represent ideas; they modify behavior. Once you frame them as operators instead of icons, the group snaps into place.
Use Process of Elimination Like a Cooldown
With Blue and Purple locked, the remaining words become dramatically easier to parse. This is where discipline matters. Don’t rush the final grouping just because the board is smaller.
By removing geometry and text operators from consideration, the remaining four are forced into a single coherent system. Connections #370 rewards patience here, not speed, and burning attempts early is the fastest way to wipe a clean run.
Each of these strategies reinforces the same core lesson: Connections isn’t a vocabulary test. It’s a systems puzzle. Once you start treating categories like mechanics instead of themes, puzzles like #370 stop feeling unfair and start feeling finely tuned.
Final Thoughts: What Made the June 15 Puzzle Unique
What ultimately set Connections #370 apart wasn’t raw difficulty, but how cleanly it enforced its ruleset. Every category asked players to commit to a single mental model and stick with it, no hedging allowed. If you tried to brute-force pattern matching instead of playing the system, the puzzle punished you immediately.
A Puzzle Built on Progressive Reads
June 15 worked best when approached like a layered tutorial. The soft hint was recognizing rigid math language in Blue, which quietly trained you to stop accepting near-matches. From there, the puzzle escalated into functional thinking rather than visual or thematic grouping.
That design made each solved category feel like unlocking a new mechanic. Once you understood how the game wanted you to think, the remaining words stopped being overwhelming and started behaving predictably.
Why the Correct Groupings Felt Earned
Each final grouping rewarded precise logic. Shapes only counted if they met strict geometric definitions. Utensils weren’t about food themes, but direct player actions in the loop. Text symbols weren’t decorations; they were operators that change behavior in a system.
The correct answers didn’t rely on trivia or obscure definitions. They relied on understanding how words function in a shared ruleset, which is why the solutions felt fair even if they took a few attempts to reach.
A Master-Class in Connections Design
What makes #370 memorable is how little RNG it contained. There were no coin-flip guesses or ambiguous overlaps once you framed each category correctly. Every wrong turn came from misreading the mechanic, not from unclear design.
That’s peak Connections. When the puzzle wipes you, it’s because you ignored the system prompt, not because the game moved the goalposts.
Final Tip for Future Puzzles
If there’s one takeaway from June 15, it’s this: stop naming things and start simulating them. Ask what the word does, how it’s used, and what rules it obeys. When you play Connections like a systems-driven game instead of a word list, even the toughest boards become manageable.
Puzzle #370 didn’t just test knowledge. It tested discipline. And that’s exactly why it stands out as one of the most satisfying solves of the month.